CE vs PMA Study Shows Little Difference In Recalls, Big Difference In Time To Market. But Surely It’s Much More Complicated Than That.

In short

In an article which is timely because of the imminent release of medical device regulatory rule changes, Eucomed has covered a piece published by Clinica late last week which itself relates to work undertaken by the Boston Consulting Group. In a study, performed by the world famous institution it would appear that CE marking of high risk devices, while it takes on average 43 months less than the granting of a U.S. PMA is virtually no less effective when measured by the number of recalls occurring during the lag between approvals.

The Eucomed article is a good read and can be found here.


The background to the “which is best” debate is that regulatory frameworks both sides of the pond are under review, with the likelihood being that many of the features of the U.S. PMA  are likely to find themselves in the new EU regs to be published in draft this week. As we’ve pointed out recently, scandals like PIP and debacles like all-metal hips and implantable cardioverter defibrillator leads, all give the regulatory bodies virtual carte blanche to rack up their requirements. And rack them up they are doing.

So, while this piece of work is a bit “after the Lord Mayor’s show”, it’s nonetheless interesting to wonder about a regime-tightening when the evidence seems to suggest the tighter regime has more disadvantages than advantages, especially for EU patients who may be delayed in their access to new technologies where previously they got them well before their U.S. counterparts.

If we’re aiming for world class devices we need to aim higher than “no recalls”

But does the data represent the full story? To our mind it doesn’t because there’s far more subtlety to it than recall data which is a pretty black and white measure of ultimate safety. What would be far more illuminating would be a comparative assessment of efficacy, because what we sense is that products that hit the streets in Europe early are often rejected in the United States, not on the basis of whether they are safe, but whether they are effective, a key difference in approach by the FDA compared with the equivalent notified body.

Does the lack of comparative efficacy data in this study devalue it then? We think it probably does, but we haven’t seen the original document so won’t comment further other than to point to the fact that the cultural and intrinsically opaque nature of things this (EU) side of the pond means problems don’t get aired like they do in the U.S and for that reason they don’t get reported as much either. If you don’t believe that, take a look at the FDA’s adverse event database, MAUDE, and we challenge you not to gasp at some of the reported incidences. Since when, in Europe, would anyone report a needle detaching from a suture. I mean, pardoning the pun, what’s the point, especially if it’s not going to show up anywhere or find its way into a formal feedback mechanism. All of which is why we’re left with the fairly blunt instrument of product recall data, when really we’d like to know how devices have worked rather than how they haven’t.

One example that rather says it all, is that of Edwards Lifesciences’ Sapien valve, about which we’ve written on many occasions. The device is readily available and from what we can gather, fairly frequently used in Europe, yet only now gaining limited indications for use in the United States. As far as we’re aware there are no reports of problems with the product, so of course it looks like Americans are being denied this brilliant new technology… a heart valve implanted without a major invasive procedure. Yet the American caution is driven by clinical studies demanded by the FDA PMA process, which indicate that the technique may only be clinically justified in a smaller patient group than that for which it is being used in Europe. Of course at this stage that’s the subject of conjecture, but if data did support that view, this would be a prime example of how recall and major safety data only doesn’t paint a full enough picture. It’s all about the subtleties and nuances of clinical experience.

All the discussion does then is emphasise one more time, as if we need it, that device regulation is, or should be, a multifactorial affair. Much as companies will use this BCG study to campaign for a lower regulatory bar, to do so is missing the point. That the PMA process has become a burdensome, lengthy, expensive and increasingly difficult journey is obvious. Without the comparatively easier route to market that is the current EU devices directive, many innovations would not have got off the ground, because there can be little doubt that companies go the EU route first in order to gain income and data to support a PMA undertaking. Beef up the European rules to PMA level and that income is lost and with it many an innovation in areas which by definition are highest risk, therefore challenging, therefore arguably in greatest need of innovation.

So… what? Well it’s a bit late to start campaigning. It’ll be what it’ll be. And taking medtech in its entirety there are clearly more products that fall into the category of supposedly “lower risk” that have suffered problems than in the higher risk group anyway… metal-on-metal hips for example were 510(k)cleared. So the focus should perhaps be on how we monitor and regulate the majority of products, and we’re wearing our most optimistic faces when we hope and expect to see a greater emphasis on post market surveillance, transparency of clinical evidence and an overhaul of the notified body system on Wednesday.

Whatever we end up with let’s remember it’s about patient well-being first, which is a fairly nebulous concept. The best way to measure a nebulous concept is to be avid collectors and processors of evidence and never fall into the trap of thinking that devices are fine as long as they don’t result in recalls.

Source: Eucomed