When Medical Device Companies Bend the Rules

What don't medical device makers do because they have figured out how to get away with not doing it?

By William A. Hyman


Back in the ancient days of 2006, I wrote a commentary in MD+DI entitled “What Wouldn’t You Do If FDA Didn’t Make You?” That piece arose from the multiple times I had heard medical device professionals and researchers suggest that they would take a faster (i.e., riskier) course if not for FDA stopping them or, perhaps, catching them.

Learn how to avoid regulatory pitfalls with early clinical involvement at MD&M Texas, May 7–8, 2014, in Fort Worth, TX.

My favorite example comes from the artificial heart arena. A presenter at a national conference said, “We would be implanting these today if FDA wasn’t making us prove that it was safe and effective.” This perhaps implied that safety and efficacy are not of fundamental concern to the developers—or at least not of sufficient concern to actually acquire proof. Perhaps this comment is at the cusp of the regulatory debate: Too much regulation stops some devices from reaching the public, but too little regulation can allow bad devices to reach the public.

I have heard and seen medical device professionals use and advocate strategies to reduce FDA scrutiny of their devices and processes, and thus improve their marketing opportunities while not necessarily improving their products. This can take the form of a do-as-little-as-possible attitude, coupled with not being forthcoming about the true nature of their medical device’s issues.

The no-510(k) rationale for device changes offers an example. This scenario arises when manufacturers make a change in their 510(k)-cleared product and must decide whether that change requires a new 510(k) and if so what kind. The use of the no-510(k) rationale can cover a spectrum from forthright, honest, and probably correct to “What do we have to argue to get away this?”

FDA regulations and guidances address this question, notably in the 1997 “Deciding When to Submit a New 510(k) for a Change in an Existing Device Guidance Document.” While nominally formulaic, the questions actually require a great deal of local interpretation. If the decision to seek a new 510(k) is made, then the manufacturer proceeds with the 510(k). If the manufacturer decides not to seek a new 510(k), it writes a memo explaining its rationalization, in effect explaining its rationale to itself because that memo does not go to FDA. FDA did propose in a 2011 draft guidance that it be notified of such a change, but that draft was withdrawn in 2012 by congressional mandate. While a new draft has not been reissued, FDA has held workshops on the issue.

Other arenas for doing less versus more arise in performance testing, biocompatibility, validation, risk management, and related design and test issues. Is the preferred strategy to do as little as possible or to apply a comprehensive good-engineering approach that while meeting the inherently minimal FDA requirements goes beyond, or proceeds independently of, these requirements to provide further assurance that the resultant device is as good as you could reasonably make it?

The PMA process also offers the opportunity for regulatory strategies that can cover a range of behaviors. One strategy that might fall into a gray zone is using a series of PMA supplements to do something you didn’t think you would get away with in a single supplement. For example, you have design A and want to revise it to design E, but you think that if you did a supplement from A to E that FDA would think the leap was too great and ask for a new PMA. The strategy then is to go from A to B, from B to C, from C to D, etc. with each change being relatively small but the end effect being design E nonetheless. With this approach, you don’t even have to actually market B, C, and D, just get them approved. A good anecdotal example of this that I once heard is that all of the functions of design E had already been built into the electronics of design A, but not all of the features had been turned on. These new functions were enabled one at a time with successive supplements. A similar strategy can be to gradually expand indications for use with the knowledge that if you had gone for the broader indications from the onset, you might have met resistance.

Are these approaches just operating within the rules, or are they cheating? Or if not outright cheating, is it what we want to happen—especially if we put on our future patient hats? In this regard, there was a recent analysis that concluded that in the cardiac device arena 77 original PMAs accumulated 5839 supplements. More than one-third of the supplements involved a change to a device’s design or materials, and in the majority of these cases, the supplements were approved without FDA requiring submission of new clinical data.

Similar issues are presented in FDA’s February 2013 draft guidance on enhancements versus recalls. Here, too, the manufacturer makes the initial decision, and if the decision is, for example, that the withdrawal of one device with replacement by another is not a recall, then FDA receives no direct notification. This kind of decision can have a strong or weak justification, with the basis being primarily patient safety, primarily business interests, or somewhere in between. Of course, business interest is a valid parameter, but perhaps not when it overshadows safety considerations. More generally, various field activities may or not be labeled a recall by the manufacturer, and if deemed to not be a recall, FDA need not be notified. Again, this may be a fair assessment or deliberate withholding of information to avoid the recall label and associated tribulations. Medical device reporting (MDR) can also challenge the system when manufacturers make nonreportable decisions, which by definition means that FDA doesn’t directly learn of the associated events.

Most rational people agree that there is a right degree of regulatory burden, and if that degree is found, manufacturers should fully comply with it without bending the rules. At some faraway end of the spectrum is the notion that FDA regulations are all burden and no value, and that this justifies finding the quickest and easiest path—even if that involves some creative avoidance. Where do you stand?

Learn how to avoid regulatory pitfalls with early clinical involvement at MD&M Texas, May 7–8, 2014, in Fort Worth, TX.

William A. Hyman is a professor emeritus in the department of biomedical engineering at Texas A&M University and adjunct professor of biomedical engineering at the Cooper Union.


Device talk Tags: