For three decades, the medical device defense bar has litigated preemption under a framework that even its authors could not agree upon.
Medtronic Inc. v. Lohr, decided by the U.S. Supreme Court in 1996, produced a fractured plurality that grafted two glosses onto the express preemption clause of the Medical Device Amendments, or MDA, at Title 21 of the U.S. Code, Section 360k(a).
But neither of those glosses finds support in the statutory text. One requires that a preempted federal requirement be “device specific” rather than general. The other permits state tort claims to proceed so long as they track duties that parallel federal requirements.
Both limitations have spawned a generation of confused, unpredictable lower court decisions. And both rest on a foundation the Supreme Court has now demolished.
In 2024, the court overruled Chevron deference in Loper Bright Enterprises v. Raimondo, holding that courts must exercise independent judgment when interpreting statutes, and may not defer to agency glosses merely because a statute is ambiguous.
Then, on June 25, the court decided Monsanto Co. v. Durnell, reading a materially identical preemption clause in the Federal Insecticide, Fungicide and Rodenticide Act in broad, textual terms that cannot be reconciled with Lohr’s atextual carveouts.
Together, these decisions strip away Lohr’s two load-bearing pillars: Chevron deference to the U.S. Food and Drug Administration’s narrowing regulation, and the notion that an agency’s product-specific approval determinations do not constitute preemptive requirements.
It is time for the Supreme Court to overrule Lohr. But vertical precedent means only the court itself can do so.
The campaign must therefore begin where all cases begin—in the district courts—with defense counsel raising and preserving the issue for the appellate path that leads to One First Street.
Lohr’s Holding and Why It Is Vulnerable
MDA Section 360k(a) provides, in relevant part, that no state “may establish or continue in effect” any requirement that “is different from, or in addition to, any requirement applicable under this Act” and that “relates to the safety or effectiveness of the device.”
The text is broad: It preempts any state requirement, including a common-law tort duty, that diverges from a federal requirement applicable under the act.
Nothing in the statutory language confines preemption to device-specific federal mandates or exempts parallel state claims. The Lohr plurality nevertheless read both limitations into the statute.
It held that clearance under Section 510(k) of the Food, Drug and Cosmetic Act—the premarket notification process for devices demonstrated to be substantially equivalent to a lawfully marketed predicate–does not impose any device-specific requirement because the process is “focused on equivalence, not safety.”
And it endorsed the proposition that a damages remedy premised on violation of duties parallel to federal requirements does not impose a requirement in addition to or different from the federal ones.
Critically, the majority arrived at these conclusions by giving substantial weight to the FDA’s own regulation, at Title 21 of the Code of Federal Regulations, Section 808.1(d), which purports to limit preemption to situations in which the FDA has established “specific counterpart regulations or ... other specific requirements applicable to a particular device.”
The court justified that weight by invoking its 1984 decision in Chevron U.S.A. Inc. v. Natural Resources Defense Council.
It reasoned that Congress gave the FDA “a unique role in determining the scope of § 360k’s pre-emptive effect” and that the statute’s ambiguity, combined with the agency’s delegated authority, provided a “sound basis” for deference.
Former Justice Sandra Day O’Connor’s partial dissent read the statute the way its text demands. She observed that Section 360k “makes no mention of a requirement of specificity.”
She further found that there is no sound basis for the majority’s restriction, and that the court “errs when it employs an agency’s narrowing construction of a statute where no such deference is warranted.”
Today, her dissent reads less like a minority position and more like a preview of the textualist majority that has come to dominate the court’s preemption and administrative law jurisprudence.
Loper Bright Removes the Foundation
When Loper Bright overruled Chevron, it did not merely tinker with the standard of review. It announced that courts “decide legal questions by applying their own judgment.”
It further asserted that the Administrative Procedure Act requires the courts to “decide all relevant questions of law” and “interpret … statutory provisions.” If there is a best reading of a statute, that reading governs—“if it is not the best, it is not permissible.”
The court said that agency interpretations may receive “respect” under its 1944 ruling in Skidmore v. Swift & Co., to the extent they are persuasive, but that they command no deference.
This holding eliminates the very rationale the Lohr majority used to adopt the FDA’s view of Section 360k(a). The Lohr plurality’s reliance on Chevron was not incidental; it was the stated basis for treating the agency’s regulatory gloss as authoritative.
Strip that away, and courts must ask independently what is the best reading of Section 360k(a). The text provides the answer: preemption of any state requirement different from or in addition to any applicable federal requirement.
Justice O’Connor reached that conclusion in her dissent. And the high court’s 2008 decision in Riegel v. Medtronic Inc. later confirmed it in the premarket approval context. Loper Bright reads far more like the O’Connor dissent than like the poorly aged Lohr plurality.
Even before Loper Bright, the court signaled that agencies possess no special competence to pronounce on preemption. In Wyeth v. Levine, the court declined in 2009 to defer to the FDA’s conclusion that state law was preempted.
And in PLIVA Inc. v. Mensing, the court stated flatly in 2011 that “we do not defer to an agency’s ultimate conclusion about whether state law should be pre-empted.”
Lohr’s invocation of Chevron for precisely that purpose was already an outlier; Loper Bright confirms it was wrong.
How Monsanto v. Durnell Reinforces the Case
If Loper Bright removes the deference pillar, the court’s recent decision in Monsanto v. Durnell removes the substantive one.
FIFRA’s preemption clause, at Title 7 of the U.S. Code, Section 136v(b), is virtually identical to Section 360k(a) of the MDA: A state “shall not impose or continue in effect any requirements for labeling or packaging in addition to or different from those required under” the act.
Writing for the court, Justice Brett Kavanaugh held that the U.S. Environmental Protection Agency’s registration and label approval determinations are requirements with preemptive force, and that a state failure-to-warn claim demanding an additional warning is therefore expressly preempted.
The court rejected the respondent’s argument that FIFRA’s general misbranding standard is the only operative federal requirement, explaining that this framing “operates at far too high a level of generality and disregards the central and comprehensive role that EPA performs in making labeling determinations.”
That is precisely the move the Lohr plurality made with respect to the MDA’s general FDCA Section 510(k) framework. Monsanto repudiates it.
The Monsanto court leaned heavily on Riegel, describing the MDA’s preemption clause as “nearly identical” to FIFRA’s, and treating Riegel’s holding—that FDA premarket approval imposes requirements under MDA Section 360k(a)—as dispositive of the parallel question under FIFRA.
In the court’s logic, if agency-specific product approval determinations constitute preemptive requirements under one statute, they must do so under the other.
The Monsanto court also confronted and rejected an argument grounded in Loper Bright itself.
The respondent contended that the EPA’s registration procedures exceeded statutory authority. The court answered that FIFRA empowers the EPA to “prescribe regulations to carry out the provisions” of the act, and directs it to register pesticides upon determining that labeling complies.
Agency determinations made under such delegated authority are requirements under the statute—exactly what the MDA does with respect to the FDA and device clearance or approval.
The implications for Lohr are direct. Riegel sidelined Section 808.1(d), the same FDA regulation Lohr embraced, describing it as adding “nothing ... but confusion” to the preemption analysis.
Monsanto confirms that agency product-specific determinations under a nearly identical statutory scheme are preemptive requirements, and that the approach of treating general statutory standards as the sole operative requirement while ignoring the substance of agency-specific determinations is analytically unsound.
A faithful, deference-free textual reading of Section 360k(a) cannot sustain Lohr’s device-specificity and parallel-claims carveouts.
A Practical Road Map That Starts in the District Courts
Lohr remains binding vertical precedent. No district court is free to disregard it, however clear the doctrinal erosion may be.
Only the Supreme Court can overrule its own decisions. The path back to the court, however, runs through the trial courts, and defense counsel must begin building it now.
First, defense counsel should frontally challenge the device-specificity and parallel-claims glosses. The argument should be grounded in the text of MDA Section 360k(a), the holdings of Riegel and Monsanto, and the reasoning of Loper Bright.
The contention is straightforward: A proper textual analysis of Section 360k(a), freed from the Chevron-driven gloss of Section 808.1(d), yields the same broad preemptive scope the court found in FIFRA’s identical language.
The word “requirement” encompasses any applicable mandate imposed under the act, including those embodied in agency determinations that a device meets statutory standards.
Second, counsel should argue that even the FDCA’s Section 510(k) process imposes device-specific federal requirements. The Section 510(k) pathway of 1996 is not the Section 510(k) pathway of 2026.
Congress strengthened the device review framework through the Safe Medical Devices Act in 1990, which added special controls providing a “reasonable assurance” of safety for Class II devices. The modern enhanced Section 510(k) process requires clinical data, performance testing, and compliance with device-specific guidance documents and recognized consensus standards.
These are substantive federal requirements directed at particular device types. The premise underlying Lohr—that Section 510(k) clearance is merely a bureaucratic gatekeeping exercise “focused on equivalence, not safety”—no longer describes the regulatory reality.
Third, counsel should expect to lose in most courts. But the goal at the trial court is preservation, not persuasion.
A district court that considers itself bound by Lohr will deny the motion. But if the argument is raised, briefed and ruled upon, it is preserved for appeal.
If it is not raised, it is waived. Defense counsel who skip the argument because they expect to lose forfeit the opportunity to build the very record the Supreme Court needs.
Fourth, counsel should choose vehicles carefully. The ideal case for certiorari involves a well-established factual record, a Section 510(k)-cleared device subject to meaningful special controls or enhanced review, a state law claim that plainly imposes requirements different from, or in addition to, federal ones, and a clear lower court ruling rejecting the preemption defense on Lohr grounds.
Multiple such cases, creating a critical mass of preserved challenges across circuits, will signal to the Supreme Court that the issue is live and ripe for resolution.
Conclusion
The doctrinal scaffolding underlying Lohr is gone. The Supreme Court overruled Chevron deference, the mechanism by which the Lohr plurality adopted the FDA’s narrowing gloss on MDA Section 360k(a). The court rejected the proposition that an agency’s product-specific approval determinations do not constitute preemptive requirements.
In Monsanto, the court read a materially identical preemption clause in precisely the broad, textual manner the MDA’s language demands. The presumption against preemption, once a headwind, has been narrowed in the court’s recent jurisprudence.
And the FDCA 510(k) process itself has evolved far beyond the “maintain the status quo” gateway the Lohr plurality described.
What remains is the opportunity to a certiorari petition. Defense counsel who raise and preserve the challenge today are laying the groundwork for the day, perhaps soon, when the court agrees to hear it. It is time.
“After Monsanto, Defense Bar Must Build Case Against Lohr,” by Terry M. Henry, Lauren E. O’Donnell, and Melanie S. Carter, was published in Law360 on August 6, 2026.