Happy 26 May 2024!

The MDR and IVDR are now in force for seven (7) years, and they are not in good shape.

I think it is safe to say that they did not deliver on their promise of a better, more streamlined and more SME and innovation friendly regulatory system.

Deadlines keep being moved, patchwork happens here and there. But it is still not a world-class regulatory system – essential parts like Eudamed are not even finished yet. Innovation is fleeing to other jurisdictions and hospitals complain about shortages of devices. This is not a good place to be, and certainly not to stay.

Legacy devices

Today is an important day under the MDR, as the era of legacy devices truly starts. The manufacturers that do not have an MDR application in the works by today will lose existing legacy device benefits.

We have now truly entered the era of legacy devices – an area in which is ever harder for the outside world to understand. Many of the non-Union jurisdictions that attach importance to CE certificates for local registration are finding it more and more complicated to understand if a certificate is actually valid or not after Regulation (EU) 2023/607 created the MDR legacy devices regime that we have today. Having legacy devices certificates is of course better than not having any certificates but I often see that manufacturers do not really know what it entails.

First, there is the issue of no significant changes. That’s right, the device’s design and intended purpose are essentially frozen until the MDR or IVDR certificate drops. That means that competitors that get their MDR or IVDR certificates earlier can outcompete you with new features that you are not able to implement yet.

Secondly, the notified bodies have until end 2027 or 2028 (depending on risk class) to finish the conformity assessment and even if they don’t by then this has no legal consequences. So there is a clear incentitve to be MDR or IVDR for the manufacturer to be CE marked as soon as possible. In that regard it is strange that notified bodies are reporting a drop in applications, both for MDR and IVDR. There is no obligation for the notified bodies to finish as soon as possible unfortunately.

Also, it is not encouraging at all that the number of application refusals is increasing, with refusal for incomplete applications and applications outside the scope of designation of the notified body accounting for the two biggest grounds of refusal. These grounds for refusal should normally be avoidable, and yet they happen more and more.

In-house produced devices

It’s also an important day under the IVDR, because all except the last of the requirements for in-house produced devices in article 5 (5) IVDR become applicable. Only the justification requirement for the use of an in-house produced device still needs to become applicable on 26 May 2028.

While the MDCG has issued guidance (MDCG 2023-1) on the subject, this guidance is really broad in some areas and shows divergence in policy (or absence thereof) between member states. It seems that most member states are just waiting with local policy until they have an idea of the likely level of compliance in their respective markets. Where member states do express themselves I’ve seen interesting things, such as for example statements that cross-border testing is not allowed under article 5 (5) IVDR – excuse me? Health institutions can provide cross-border medical services to patients in the EU but not testing services? That would be very much at odds with the internal market.

The health institutions, for their part, have a hard job complying – especially the hosptial labs that are invariably under-resourced to do the extra work required under article 5 (5) IVDR.

So we seem to be entering into a kind of wild west phase for in-house produced devices. We have rules but scattered compliance and little enforcement.

A European plan economy for medical devices

We have a legislative proposal in the works that will amend the MDR and IVDR with a duty in the new article 10a MDR and IVDR to report interruptions and discontinuations of supply of a device. The proposal will also provide for a mechanism to phase-in Eudamed by module and move IVDR legacy devices deadlines. See here for Q&A.

The Council is expected to approve this on 30 May after which it will be published and enter into force somewhere in June most likely. The provision about supply interruption will enter into force six months after entry into force of the amendment, so this autumn. Manufacturers should be ready for this.

Although the proposal started out as targeting only interruptions of supply of a device but the version voted on by the Parliament additionally included discontinuation of supply of a device. As the proposal currently stands:

“a manufacturer [that] anticipates an interruption or a discontinuation of the supply of a device, other than a custom-made device, and where it is reasonably foreseeable that such interruption or discontinuation could result in serious harm or a risk of serious harm to patients or public health in one or more Member States, the manufacturer shall inform the competent authority of the Member State where it or its authorised representative is established, as well as the economic operators, health institutions and healthcare professionals to whom it directly supplies the device, of the anticipated interruption or

The information referred to in the first subparagraph shall, other than in exceptional circumstances, be provided at least 6 months before the anticipated interruption or discontinuation. The manufacturer shall specify the reasons for the interruption or discontinuation in the information provided to the competent authority.”

There will also be additional obligations for economic operators: all importers, distributors and other economic operators involved (e.g. procedure pack or system composers) economic operators who have received the information from the manufacturer or from another economic operator in the supply chain shall, without undue delay, inform any other economic operators, health institutions and healthcare professionals to whom they directly supply the device, of the anticipated interruption or discontinuation.

This means that manufacturers will need to do an impact assessment with an at least 6 months horizon at least that looks at individual Member States in the Union if they plan to discontinue or temporarily stop supply of a CE marked device. And then tell everybody and their mother about it if this may create issues in one or more member states. Distributors and importers in the Union need to establish processes to inform the supply chain downstream as well.

I understand that the MDCG is working on guidance about how to apply this provision, which is a good idea as it is very vague. My own legal experience with broadly and vaguely drafted security of supply information obligations is not good.

I am litigiating several cases in relaton to medicines where companies were find for not complying with quite unclear and broadly worded similar requirements, I hope we can avoid this in devices. In the meantime these rules for medicines are being amended to become much more sophisticated in the review of EU pharmaceutical law. Why has the legisalator not looked at the experience gained in this sector? Your guess is as good as mine.

What’s next?

The Commission has announced in January when it made the propsal that is now nearing the end of the legislative pipeline that it will pursue further amendments to the MDR and IVDR to fix ongoing problems. The European Parliament also announced when it had voted for the latest amendment on 30 April 2024 that this would not be the end of amendments to the MDR and IVDR.

One of the most recent steps is that Peter Liese, member of the European Parliament and long since involved with the MDR and the IVDR, has forwarded a proposal for Parliament suggested changes to the European Commission, as was made public on a press conference on 17 May. I had the honour to be retained by Mr. Liese’s office to serve as external expert for developing this proposal.

The European Parliament does not have the right of initiative for amendments to legislation so this is not an actual pending legislative proposal now. However, Parliament can of course seek support within the Commission for this idea so the Commission may decide to table a proposed amendment for the Council and the Parliament in a legislative procedure that includes the Parliament’s ideas because it increases the chance of success and speed of the proposal. I understand this is exactly what Mr Liese has been doing.

As said above, the Commission is also working on its own plans as it announced back in January:

“The Commission will already start in 2024 its preparatory work for a targeted evaluation of the legislation on medical devices. The evaluation should assess how the legislation is affecting the availability of devices, in particular for devices with specific characteristics (e.g. paediatrics, orphan, innovative devices). In the assessment, special attention may also be given to costs and administrative burdens stemming from the implementation of legislation, especially for SMEs.”

The question is now what will happen with the Parliament elections and Commission reshuffle coming up. The Commission will have to determine what actions might still be kicked off under the current Commission, and what exact proposal it will want to make at what time.

In any event, more and more developments seem to be converging that will likely lead to another revisiting of the MDR in the short term. And this will go beyond deadline amendments as nobody seems to want to amend the deadlines further.

For me personally a number of things need amending:

  • The EU system needs centralization of governance because the current decentralised approach does not work. There is too much difference between the way member states apply the MDR and IVDR and how they instruct notified bodies to apply these (as I notice in practice each day – try and obtain an article 59 MDR derogation for the same device in a number of member states, the diversity of approach and results are ‘interesting’).
  • Notified bodies can improve a lot in good administrative practice (although there are scattered initiatives to improve underway), while they are wielding delegated state power of control over market access. I notice in practice that member states control them rather poorly on this point and notified bodies are not able to recognise and correct their (procedural) mistakes quickly, resulting in devastating consequences for the manufacturer concerned. Let’s not forget that good administrative practice in exercise of government power is an EU constitutional right – we should expect the same level of diligence from notified bodies procedurally as we expect from authorities. Finally, there is often no meaningful appeal to a decision of a notified body after internal appeal at the notified body. It is often unclear in national law and not addressed at MDR/IVDR level if a court is competent to review a notified body decision. Legal review of exercise of state power is also an EU fundamental right. Yet, the MDR only provides for an explicit appeal pathway in case of a classification decision.
  • The approval system should be (much) more supportive of innovation and specific groups of devices (such as orphan devices). In other jurisdictions with which the Union is in regulatory competition special procedures exist for such devices, so this not be a question for the Union. Manufacturers (even EU ones) with the most innovative devices are more and more abandoning the Union as first jurisdiction of launch because the system is too slow, too costly and too unpredictable.
  • There is too much repeating bureaucracy under the MDR and IVDR. Especially the 5 years recertification cycle needs revisiting in view of the much increased PMS obligations under the MDR and IVDR. If periodic re-approval is not needed for medicines, why should it be needed for devices?
  • Much more attention is needed to the way horizontal legislation dovetails with the MDR and IVDR, especially where it comes to design changes that may need to be implemented for legacy devices as a result (e.g. Batteries Regulation, AI Regulation). Such changes are not automatically recognised as non-significant changes mind you and may therefore invalidate a legacy device certificate. More in general the regulatory lasagna created by stacking several layers of horizontal regulation on vertical regulation does not help for compliance. More in general double CE marking (as imposed by the AI Regulation for example) is just a bad idea. I understand that the MDCG and the AI Office under the soon to enter into force AI Regulation will do their best to synch policy and provide coherent guidance. But this should not be needed from a lawmaking perspective. It is infinitively unproductive to make law that is prohibitively complex to comply with (more about the AI Regulation on this blog in this respect when it enters into force).
  • Medical devices policy makers and market surveillance should be properly resourced, both on EU and national level. For a reason that entirely keep escaping me the level of resources devoted to policy and market surveillance in medicines is much much better while healthcare simply cannot function without medical devices and IVDs.

This was my personal wish list, and many other wish lists are circulating. Things are moving and positions converging. Now it’s time to fix the MDR and IVDR.

Happy birthday MDR and IVDR – I’m curious what the next one looks like!

In any event the new third edition of The Enriched MDR and IVDR will have been published by then, one of the reasons that this blog has received less attention from me than usual. Many people have told me that they miss it, so I will do my best to post more often again.

Navigate through our knowledgebase

Related articles


The German angle

After having contributed to the improvement of medical devices legislation in many ways over the years (most recently by helping the European Parliament to come up with an initiative for targeted amendments…

Read more


A case of so-called fiscal neutrality

Sometimes you come across cases that violate Mandalorian Creed: “One does not speak unless one knows.”. This happened to me last week when I read the Dutch Supreme Court’s judgment in a…

Read more


Can we fix / improve the MDR and the IVDR?

Or in other words that I’ve asked on this blog before: can the maker repair what he makes? This blog will argue that he can and he should. It still happens to…

Read more