Federal preemption in the drug and medical device world is a game of categories. Express preemption versus implied. Conflict preemption versus field preemption. Drugs versus devices, generic versus branded, premarket approved versus not. These categories make a difference, and if you can place your controversy into the correct boxes, you can usually determine how to analyze whether federal law preempts state-law claims. Sometimes you can even predict the result, or at least predict what the correct result should be. We say “should” because there is discretion involved, and also some ambiguity in the applicable law. Moreover, preemption is like hockey: Many Americans can discuss it with some familiarity, but a much smaller number have studied all the nuances of the game.
From time to time a preemption order thinks outside the box, and that is what happened in Williams v. Zimmer US, Inc., No. 5:14-cv-468, 2015 U.S. Dist. LEXIS 91238 (E.D.N.C. July 14, 2015). The case involved a bizarre situation where a surgeon took a medical device home, and “in his own garage, [he] used nonmedical tools purchased from a home improvement store to bend a 5.5 mm chrome rod into a ‘U’ shape.” Id. at *4. Facts like these make us scratch our heads, but putting the weird facts aside, the device was cleared through the 510k premarket notification process. As our readers know, under Lohr and Riegel, there generally is no express preemption under the Medical Device Amendments for 510k-cleared devices, so our knee-jerk reaction when a 510k device is involved is to put that case into a “no preemption” box.
But don’t forget about implied preemption, which is where this district judge went with this case. Our do-it-yourself doctor used his Home Depot version of the device to treat a patient, who later sued alleging injuries. But she couched her complaint entirely in terms of violations of the FDCA.
That had two dramatic consequences. First, the defendants removed the case to federal court, and the district court ruled that it had subject matter jurisdiction because the case arose under federal law, i.e., the court had federal question jurisdiction. Id. at **6-7. This is exceptional because the vast majority of drug and medical device cases in federal court are there under diversity jurisdiction, even though the industry is subject to intense federal regulation. An opinion finding federal question jurisdiction in a medical device case is a welcome development, and this plaintiff seems to have pleaded straight into it.
Second, because the plaintiffs placed the FDCA and alleged violations thereof at the center of their complaint, the district court ruled that federal law preempted the state-law claims under Buckman and dismissed the case. Id. at **11-17. This is not an application of implied conflict preemption that we often see. Nor is it routine to apply Buckman outside the context of fraud-on-the-FDA claims, although there are multiple cases where courts have done that. This district court did so too, and its reasoning was absolutely correct. Again, the court applied implied conflict preemption, under which “state law may conflict with federal law either because ‘compliance with both state and federal law is impossible’ or because the state law impedes a federal purpose. Id. at *11. In Buckman, the Supreme Court invoked implied conflict preemption to hold that fraud-on-the-FDA claims were preempted. As this district court saw it,
Buckman did leave room for a small range of state law claims based on the FDCA, namely those state-law claims that would rely on traditional state tort law independent of the FDCA. Put in other terms, “a state law claim only endures if it manages to incorporate, but not depend entirely upon, and FDCA violation and is premised on conduct that would give rise to liability under traditional common law principles.” If, however, the defendant would not be liable “but for the FDCA, ‘then the plaintiff is effectively suing for a violation of the FDCA (no matter how the plaintiff labels the claim), and the plaintiff’s claim is thus impliedly preempted under Buckman.’”
Id. at **12-13 (citations omitted). This is a very good gloss on Buckman, although we would read the Supreme Court’s preemption ruling even more broadly. But we will not dwell on that here because the district court’s application of Buckman to dismiss this case was bang on. First, the court ruled that it is now settled that Buckman’s holding is not limited to fraud on the FDA. Citing multiple examples, the court reiterated that federal law impliedly preempts any claim where the plaintiff is effectively suing for violations of the FDCA, including claims pleaded as breach of warranty, negligence per se, design defect, and failure to warn. Id. at **14-15.
Here, the plaintiffs alleged a number of purportedly state-law claims, but all were premised on either (1) off-label use of the device or (2) failure to advise that the FDA had not yet approved or cleared other devices used in the surgery: These claims were all preempted because “Buckman is not limited to fraud-on-the-FDA claims” and because “the present claims depend entirely upon alleged violations of the FDCA.” Id. at *16. As we know from Buckman and many other cases, there is no private right of action for violations of the FDCA.
We also like how this district court understood that express preemption and implied preemption are two different things. The plaintiffs argued that federal law did not preempt their claims because the MDA’s express preemption provision did not apply. Id. at *13. Fair enough, but that does not foreclose implied preemption. As the district court stated, “Buckman dispensed with the notion that an express preemption provision prevents the application of [implied] conflict preemption.” Id. at *17. We will acknowledge that the plaintiffs’ argument was not outlandish. The Supreme Court in Wyeth v. Levine did give some credence to the idea that the lack of an express preemption clause could suggest a Congressional intent that there should be no preemption of any kind. We think that is wrong. The Supreme Court has ruled on multiple occasions—both before and after Levine—that implied preemption operates independently from and without reference to express preemption. That is the law and that is the way it ought to be.
Finally, the district court dismissed the case with prejudice because any amendment to the complaint would be futile. We routinely express our dissatisfaction with orders granting leave to amend when circumstances do not warrant it, so this part of the ruling is welcome news too, mainly because the plaintiffs will not get to fish around in expensive discovery. The main story though is Buckman-style preemption and this district court’s exceptionally clear minded application of implied preemption to the plaintiffs’ allegations.