Next Decade To-Do: Enforce Preemption for Class II Devices with Special Controls

The common assumption is that FDA premarket approval of a Class III device is a necessary prerequisite for express federal preemption of state law design defect claims.

But that assumption is wrong. FDA adoption of special controls for Class II devices can also expressly preempt design defect claims even though those devices are not subject to premarket approval and are cleared using the §510(k) notice process.

The operative word is “can.” Some special controls are more likely to preempt than others. In the broadest terms, the more specific the control, the more likely it is to preempt state tort law. Controls expressed in regulations are more likely to preempt than controls in the form of a FDA guidance adopted by a regulation. And courts should be more likely to recognize the preemptive effect of a control when the FDA itself has done so. At this point, though, these are mere likelihoods, not certainties.

But even if the path to express preemption is muddied, it is not impassable. The time has come to consider express federal preemption anytime a state tort suit is brought challenging a Class II device whose classifying regulation adopts a special control. Success in a strong case would clear the way for others to follow.


There remains a gap in the kind of device regulations whose preemptive effect the U.S. Supreme Court has considered. While the Court has soundly precluded preemption for Class I devices as a whole and soundly upheld design preemption for Class III devices, in the middle fall Class II devices, some subject to special controls and others not. Assuming there is no preemption for any Class II device overlooks the law that governs devices which are subject to “special controls.”

The FDA regulates medical devices in three different ways, each roughly corresponding to the degree of risk the FDA believes the device poses. It places the least risky devices in Class I and allows them to be marketed without any premarket FDA review. More risky devices go in Class II and may be subject to special controls. In any event, they generally may be marketed only after clearance using the §510(k) notice process. That process requires a showing of safety and equivalence to a device already on the market. Finally, the most risky devices are placed in Class III and may only be sold after being approved based on independent evidence of their safety and effectiveness.

The Food, Drug, and Cosmetic Act provides that a state may not “establish or continue in effect” any “requirement” which is “different from, or in addition to, any requirement applicable under this chapter to the device….”[1] This broad language would appear to foreclose a state tort suit against any regulated device now that it has been made clear that the word “requirement” includes state tort law rules.[2]

The FDA, however, has attempted to narrow the statute by construing “requirement applicable … to the device” to apply only where the FDA “has established specific counterpart regulations, or there are other specific requirements applicable to a particular device under the act….”[3] Construing this language, the U.S. Supreme Court has said that general FDA regulations – regulations applicable to all devices – do not count as “requirements.”[4] Thus there is no express preemption for Class I devices or Class II devices not subject to special controls. And, at the other end, the Court has said that FDA approval of Class III devices does give rise to requirements “applicable to a particular device” that preempts state design defect law.[5]

The category that the Court has not addressed – and which bears consideration by defendants in device litigation – falls in the middle: Class II devices, which are subject to special controls. The 1976 Act establishes the framework for classification and the imposition of special controls. It instructs the FDA to convene panels of medical experts, whose qualifications are dictated by statute, to classify medical devices. The panels consider available medical literature, review injury reports sent to the FDA, hold hearings, and then decide how devices are to be classified in order to provide “reasonable assurance of safety and effectiveness.” 21 U.S.C. §360c(a)(1). A device or category of devices can only be placed in Class II if the panel concludes it does not present “a potential unreasonable risk of illness or injury” that would require it to be placed in Class III. 21 U.S.C. §360c(a)(1)(C)(ii)(II).

When the FDA adopts a regulation that puts a device or category of devices in Class II, it may or may not impose “special controls” that, under the statute, can include “performance standards … development and dissemination of guidelines (including guidelines for the submission of clinical data in [510(k) notifications]), recommendations, and other such actions as the Secretary deems necessary” to provide reasonable assurance of safety and effectiveness.[6]

The question then arises as to whether “special controls” included in the FDA regulation, which classifies the device or category of devices, are “requirements” that trigger express federal preemption. If they are, then the fact that individual Class II devices are cleared using §510(k) notice rather than premarket approval is irrelevant. The 510(k) process ensures compliance with special controls for that category of devices, and that is all that is needed for preemption.


In the wake of claims that tampons caused “toxic shock” in women, the FDA’s Obstetrics-Gynecology and Radiologic Devices Panel recommended, and the FDA adopted in 1982, 21 C.F.R. § 801.430, “User labeling for menstrual tampons.” The FDA kept tampons in Class II but required this language:

ATTENTION: Tampons are associated with Toxic Shock Syndrome (TSS). TSS is a rare but serious disease that may cause death. Read and save the enclosed information.

The FDA also said manufacturers had to address certain subjects, such as the warning signs of toxic shock syndrome and how to treat it.[7] In 1989, the FDA amended the regulation to require testing and to standardize absorbency labeling.[8] The same regulation also required a description of toxic shock risk and a warning to use tampons with minimum absorbency.[9]

In Papike v. Tambrands Inc.,[10] the Ninth Circuit held that the tampon regulation preempted not only failure to warn claims, but also design defect claims based on a consumer expectations theory and express warranty claims. According to the court, the FDA “regulation mandating the specific substantive content of the TSS warnings on tampon boxes and/or tampon package inserts” in 21 C.F.R. § 801.430 was “not only device-specific (tampons), but also disease-specific (TSS).”[11] The Ninth Circuit also said that the FDA’s statement that regulatory standards could be met in other ways, i.e. that manufacturers could draft their own warning language, did not defeat preemption.[12]

The FDA recognizes that its special controls can have preemptive effect. A 1999 Presidential Executive Order requires that federal agencies issue a “federalism summary impact” statement when a regulation preempts state law.[13] After the Supreme Court squarely held in Riegel v. Medtronic[14] that state tort law rules can be preempted, the FDA began issuing federalism statements when it designated special control guidance. There are at least 26 final rules beginning in 2008 that contain these statements.

The earliest example came with the final adoption of a condom regulation in 2008. That regulation, 21 C.F.R. §884.5300, does not expressly impose standards but instead provides: “The guidance document entitled ‘Class II Special Controls Guidance Document: Labeling for Natural Rubber Latex Condoms Classified Under 21 CFR 884.5300’ will serve as the special control.”

The guidance, in turn, requires that labeling specify an expiration date, give the natural rubber latex warning, and include adequate directions for use. It also recommends that labeling specify that condoms reduce but do not eliminate the risk of pregnancy and sexually transmitted diseases. The labels must list the failure rate and the condom’s relative lack of effectiveness in preventing certain diseases. Examples of acceptable statements are given, e.g., “Latex condoms do not completely eliminate the risks of pregnancy and sexually transmitted infections (STIs).” And six specific directions for use are given. See

In describing the preemptive effect of the regulation, the FDA acknowledged that normally a “guidance” need not be followed and so cannot be considered a “requirement.” But it distinguished a “guidance” that constitutes a special control:

Unlike a regular guidance, which imposes no requirements, where a guidance document has been designated as a special control by a rule, manufacturers must address the issues identified in the guidance, either by following the recommendations in the guidance or by some other means that provides equivalent assurances of safety and effectiveness. At the same time, establishing a guidance document as a special control affords greater flexibility than a rule mandating specific labeling language and can facilitate updating labeling as new scientific information becomes available because the special control permits manufacturers to use any labeling that affords equivalent assurances of safety and effectiveness for latex condoms.[15]

It identified the issues requiring the special control as unintended pregnancy, sexually-transmitted infections, and incorrect use.

In Rasheed v. Church & Dwight Co.,[16] the plaintiff sued the manufacturer of Trojan condoms after condom use was followed by a severe reaction. Because the guidance addressed the potential for allergic reactions to natural rubber latex or the condom lubricant, the Court, in an alternate ruling, held that plaintiff’s failure to warn claim was preempted. In so holding, it relied on the applicable federal regulation and the tampon precedents.[17]


Either case law or FDA federalism statements have now recognized that a special control for a Class II device can preempt state law claims with respect to a wide variety of devices. These include not only tampons and condoms but also latex gloves,[18] contact lenses,[19] angioplasty catheters,[20] wound dressing,[21] tissue adhesive with wound closure device,[22] a hemorrhoid prevention pressure wedge,[23] and electrical stimulation devices.[24] But courts have rejected preemption claims with respect to bone cement[25] and IVC filters.[26]

Some of these regulations have prescribed labeling,[27] but most have not. Some have identified special control features in the regulation itself,[28] but others have simply adopted a guidance as a special control.[29]

A few things can be said, however, about the cases in which courts have refused to give preemptive effect to a special control. Both of them concerned regulations adopted before 2008; meaning they did not consider FDA federalism statements, either with respect to the device in issue or in general.[30] One simply dismissed the idea that a special control guidance could impose any “requirements” and, confusing the issues, thought that it mattered that the device was cleared using §510(k).[31] Another questioned whether a guidance, even one required by a classifying regulation, was sufficiently device-specific, and similarly opined that §510(k) review was not the same a premarket approval.[32]

In addition, some courts have distinguished among state tort claims and refused, for example, to give a labeling requirement preemptive effect with respect to a design defect claim based on a risk-benefit analysis.[33]

Some of these criticisms, however, appear to be misplaced.

First of all, the continued validity of the FDA’s narrowing regulation, 21 C.F.R. § 808.1(d), is subject to question. As far back as 1996, four justices – Sandra Day O’Connor, Antonin Scalia, Clarence Thomas, and Chief Justice William H. Rehnquist – doubted that the FDA had the authority to narrow the statute.[34] And, in an opinion as a Court of Appeals judge, Justice Neil Gorsuch noted that the regulation conflicts with “the statute’s literal language,” something he called “no small mystery.”[35]

Second, the language of the regulation when read carefully does not require that the “specific counterpart regulations” be specific to an individual manufacturer’s device. In fact the Supreme Court case that first limited express preemption, Medtronic v. Lohr,[36] referred to the kind of federal regulations to be protected from state interference as regulations “regarding a specific device or field of device.”[37]

Third, there is no reason for courts to be dismissive of §510(k) clearance when that clearance rests on medical panel classification of a device or “field of device” in Class II, for that constitutes a finding that the device does not present an “unreasonable risk of illness or injury.” There was no such panel action or classification at issue in Lohr, the case mistakenly relied upon as being typical of §510(k) review and used to defeat preemption for §510(k) cleared devices.

To the contrary, courts should be particularly deferential to Class II designations and special controls found in regulations adopted only after prior notice and opportunity for comment in the Federal Register. That is a level of public scrutiny that “approved” devices do not get. And the deference would be justified: The likelihood that a Class II device cleared using §510(k) will result in a serious recall is lower than the likelihood that a Class III premarket approved device will result in a serious recall.[38]

Fourth, affording express preemptive effect to a special control would be consistent with the law of conflict preemption. Conflict preemption arises where the FDA’s “special permission or assistance” is required to market a device.[39] A manufacturer cannot make a “significant change” to a device without first obtaining FDA clearance.[40] If a question should arise as to whether a change is “significant,” there could be no better evidence of significance than the existence of a special control guidance that has identified the design characteristic being changed, e.g., the material used, as something that must be disclosed in a §510(k) application.

For all these reasons, express preemption should be considered in any defense of a Class II medical device for which the FDA has adopted a regulation imposing special controls.

[1] 21 U.S.C. §360(k).

[2] See Riegel v. Medtronic, Inc., 552 U.S. 312, 324-25 (2008); id. at 327-28.

[3] 21 C.F.R. §808.1(d).

[4] Medtronic v. Lohr, 518 U.S. 470, 498-501 (1996).

[5] Riegel, 522 U.S. at 322-23.

[6] 21 U.S.C. §360c(a)(1)(B).

[7] 47 Fed. Reg. 26982-01 (June 22, 1982).

[8] 54 Fed. Reg. 43766 (Oct. 26, 1989).

[9] Id.

[10] 107 F.3d 737, 742-44 (9th Cir. 1997).

[11] Id. at 740.

[12] Id. at 741. See also Murphy v. Playtex Family Prods. Corp., 69 F. App’x. 140, 143-44 (4th Cir. 2003) (preemption of failure to warn and consumer expectations design defect claim).

[13]  64 Fed. Reg. 43255 (Aug. 4, 1999).

[14] 552 U.S. 312 (2008).

[15] 73 Fed. Reg. 66522-01, 66524 (Nov. 10, 2008) (emphasis added).

[16] 2012 WL 262619, *1-2 (E.D. Tex. 2012).

[17] Id. at 7-8.

[18] 21 C.F.R. § 801.437; Busch v. Ansell Perry, Inc., 2005 WL 877805, at *2-3 (W.D. Ky. Mar. 8, 2005); Morgan v. Abco Dealers, Inc., 2007 WL 4358392, at *6-7 (S.D. N.Y. Dec. 11, 2007).

[19] 21 C.F.R. § 886.5918; Tuttle v. CIBA Vision Corp, 2007 WL 677134, at *2 (D. Utah Mar. 1, 2007); Degelman v. Advanced Medical Optics, Inc., 659 F.3d 835, 841-42 (9th Cir. 2011), vacated due to settlement, 699 F.3d 1103 (9th Cir. 2012).

[20] 21 C.F.R. § 870.5100; 75 Fed. Reg. 54637-01 (Sept. 8, 2010).

[21] 21 C.F.R. § 878.4015; 74 Fed. Reg. 53269-01 (Oct. 16, 2009).

[22] 21 C.F.R. § 878.4011; 75 Fed. Reg. 68972-02 (Nov. 10, 2010) (guidance-imposed requirements “to address each health risk identified with these devices”).

[23] 21 C.F.R. § 884.5200; 76 Fed. Reg. 21237-01 (April 15, 2011).

[24] 81 Fed. Reg. 24386-01 (April 25, 2016) (proposed 21 C.F.R. § 882.5235 and 21 C.F.R. § 895.105).

[25] 21 C.F.R. §888.3027; Thompson v. DePuy Orthopaedics, Inc., 2015 WL 7888387, *8-10 (S.D. Ohio Dec. 4, 2015).

[26] 21 C.F.R. § 870.3375; In re: Bard IVC Filters Products Liability Litigation, 2017 WL 5625547 (D. Ariz. Nov. 22, 2017).

[27] See 21 C.F.R. § 801.430 (tampons); n. 4 (latex gloves), supra.

[28] See n. 9 (hemorrhoid prevention pressure wedge) and n.10 (electrical stimulation devices), supra.

[29] See n.5 (contact lens); n.6 (angioplasty catheters); n.7 (wound dressing); n.8 (tissue adhesive with wound closure device); n. 11 (bone cement); n.12 (IVC filters), supra.

[30] Thompson, 2015 WL 7888387 at *8-10; In re: Bard IVC Filters, 2017 WL 5625547 at *5-9.

[31] Thompson, supra, n. 11, at *8-9.

[32] Bard IVC Filters, supra, n. 12, at *8-11.

[33] See Busch, supra n. 4 at *1-4.

[34] Lohr, 518 U.S. at 513-14.

[35] Caplinger v. Medtronic, Inc., 784 F.3d 1335, 1339-40 (10th Cir. 2015) (Gorsuch. J.).

[36] 518 U.S. 470, 498-501 (1996).

[37] Id. at 501 (emphasis added).

[38] Jeffrey K. Shapiro, Substantial Equivalence Premarket Review: The Right Approach for Most Medical Devices, 69 Food & Drug L.J. 365, 389-90 (2014).

[39] PLIVA, Inc. v. Mensing, 131 S. Ct. 2567, 2580-81 (2011).

[40] See 21 U.S.C. § 360c(i)(1)(B); 21 C.F.R. § 807.81(a)(3); 21 C.F.R. § 807.100(a)(5).See also Mensing, 131 S. Ct. at 2576-77; Mutual Pharm. Co. v. Bartlett, 133 S. Ct. 2466 (2013); Yates v. Ortho-McNeil Pharm., Inc., 808 F. 3d 281, 298 (6th Cir. 2015).