510(k) Evidence: A View of the Universe Outside of Pelvic Mesh

When it comes to 510(k) evidence, there are two distinct universes. This article will not address the universe of pelvic mesh cases other than to point out that, even when regulatory compliance is a defense, the courts have not allowed defendants to tell the jury that the FDA has placed a device in Class II and cleared it using 510(k), the premarketing notice procedure in §510(k) of the original food and drug law, now 21 U.S.C. §360(k). In 2013, the MDL court declared that 510(k) evidence was inadmissible because it does not “go to whether the product is safe and effective.” Cisson v. C.R. Bard, Inc., 86 F. Supp. 3d 510, 514-15 (S.D.W. Va. 2015) (citing 2013 pre-trial ruling), aff’d, Cisson v. C.R. Bard, Inc. (In re C.R. Bard, Inc., Pelvic Repair Sys. Prod. Liab. Litig., MDL No. 2187), 810 F.3d 913, 919-23 (4th Cir. 2016). On appeal, the Fourth Circuit affirmed with the statement that “the clear weight of persuasive and controlling authority favors a finding that the 510(k) procedure is of little or no evidentiary value.” 810 F.3d at 920. And that view has subsequently prevailed in those cases.

This article will address a second and quite different universe of cases that do not deal with pelvic mesh. Butler Snow did not represent the defendant in Cisson, but our firm is active in the MDL and elsewhere on behalf of Johnson & Johnson and its subsidiaries in pelvic mesh cases. Because in Pro Te, Butler Snow does not write about the firm’s ongoing cases, this article will not address the pelvic mesh decisions but will instead describe the universe of all other cases that discuss the admission of 510(k) evidence. As it turns out, the Fourth Circuit’s “weight of authority” declaration is both incorrect and misleading.   

510(k)’s place in FDA regulation of devices

The 510(k) clearance process is the workhorse of medical device regulation.[1] Roughly 98% of all devices for which FDA marketing permission is required[2] have been deemed safe enough to be cleared using that process.

In 1976, Congress directed the FDA to engage in a sort of regulatory triage. Triage sorts patients according to the seriousness of their injuries and gives them different levels of care. The FDA’s statutory scheme sorts devices according to the seriousness of the risks they present and requires different levels of regulatory review.[3]

Congress told the FDA to use medical panels to assign device types to one of three categories and specified a level of review for each. Generally speaking, the lowest risk devices, Class I, were considered safe without any need for FDA premarket review. Devices that presented a moderate risk, Class II, could be “cleared” if a manufacturer’s 510(k) notice established that the device was at least equivalent in safety and effectiveness to predicate Class II devices previously cleared. Device types in the highest risk category, including those that may “present an unreasonable risk of illness and injury,” were to be assigned to Class III and could not be sold unless the FDA “approved” them based on independent evidence of safety and effectiveness.[4]

“Viewing the FDCA [Food, Drug and Cosmetic Act] as a whole, it is evident that one of the Act’s core objectives is to ensure that any product regulated by the FDA is ‘safe’ and ‘effective’ for its intended use. . . . Regardless of which category the FDA chooses, there must be a ‘reasonable assurance of the safety and effectiveness of the device.’” Food & Drug Admin. v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 133-34 (2000).

The FDA’s own Guidance Document about the 510(k) process also refutes any misunderstanding that 510(k) review does not address safety: “[T]he principles of safety and effectiveness underlie the substantial equivalence determination in every 510(k) review,” because “[s]afety and effectiveness factor into both parts [assessment of intended use and comparison to the predicate device] of the FDA’s review.”[5]

The 510(k) notice has to contain evidence to support the claim that the device is as safe and effective as the predicate device, and also “any adverse safety and effectiveness information.”[6] The manufacturer must explain how the device functions and describe the design, material used and physical properties. It must also state the intended use of the device and any testing or performance data.[7]

The FDA reviews the notice and prepares a 510(k) Reviewer Memorandum, sometimes called an add-to-file review, screening checklist, revised review, modification memorandum, SE determination or memorandum to the record, which describes the FDA’s analysis of the notice. The FDA may or may not ask for additional information. If clearance is allowed, the FDA issues a clearance letter, which confirms the substantial equivalence determination and permits sale of the device.[8]

The triage system appears to work. For example, one estimate is that while 510(k)-cleared devices constitute 98% of all devices for which marketing permission is required, in a study of devices approved and cleared between 2005–2009, cleared devices made up only 71% of all serious recalls. By contrast, 2% of approved devices result in 19% of all serious recalls.[9] These numbers tell a positive story about the effectiveness and efficiency of classification and clearance despite the fact that the 510(k) notice and the FDA’s clearance of a device may involve far less information and fewer agency resources than approval of a Class III device.

This presents a paradox, but a sensible one. The Class II cleared devices that receive less FDA scrutiny are, on average, safer than the Class III devices that the FDA has more formally approved. But that is because the Class II devices are, by definition, devices of a type that a medical panel has determined to present a lower risk. If the devices presented a “potential unreasonable risk of illness or injury,” 21 U.S.C. §360c(a)(1)(C), they would have to be in Class III, not Class II.

Again, the concept of triage is helpful. Some emergency room patients are sent home with medicine while others go to the operating room. No one would say that the patient sent to surgery was healthier just because more intensive treatment was given. It similarly makes no sense to judge the safety of a device by the intensity with which the FDA scrutinizes it, when the level of scrutiny is commensurate with the level of risk the device type presents.

There is an exception to this scheme that has played an unfortunate role in the case law. Congress in 1976 correctly anticipated that it would take the FDA a long time to convene enough medical panels to classify the universe of medical devices. As a transitional mechanism for device types that had not yet been classified, it allowed clearance based on equivalence to a device that was on the market in 1976. Clearance was allowed even if the device was implantable and thus was temporarily assumed to be a Class III device.

In 1982, the FDA cleared an implantable device based on equivalence to a pre-1976 device of a type that had never been classified. When the device was before the U.S. Supreme Court on a preemption question, the Court said that the FDA’s focus was on “equivalence, not safety” and there had never been any formal medical review. Medtronic, Inc. v. Lohr, 518 U.S. 470, 492-94 (1996). The Supreme Court opinion wrongly assumed that the use of 510(k) in that case was typical of 510(k) generally, which the Court said was not “intended to do anything more than maintain the status quo.” Id. at 494. The use of 510(k) to clear classes that were assumed to be high-risk devices subsequently generated significant criticism.[10]

But none of the context surrounding the pacemaker in Lohr is relevant for the Class II devices for which 510(k) was designed: device types that have received formal medical review.[11] As a matter of federal law, 21 U.S.C. § 360c(a)(1)(B), there is “reasonable assurance” of the safety and effectiveness of those devices, and they do not present an unreasonable risk that would put them in Class III.

Reasons for admitting 510(k) evidence

Because the FDA puts high risk devices in Class III, most litigation has involved those riskier devices. There has, however, been litigation over the performance of Class II devices cleared using 510(k), and evidence has been admitted for a number of reasons. Generally, courts that have admitted the evidence have done so with instructions that FDA clearance was not binding on the jury but could be considered.

For defendants:

For example, in Hegna v. E.I. du Pont de Nemours and Co., 806 F. Supp. 822, 829-30 & n.14 (D. Minn. 1992), a suit against the supplier of material used by a medical device maker, the court held that FDA regulation of the Class II device was not an absolute defense but was “one factor” that the jury could take into account.

In Musgrave v. Breg, Inc., No. 2:09-CV-01029, 2011 WL 4620767, at *3 (S.D. Ohio Oct. 3, 2011), a Class II pain pump case, the court held the plaintiffs could not keep the jury from hearing that the FDA cleared a general indication for use and that the FDA did not voice concerns about the pain pumps. The probative value of this evidence was said to outweigh the danger of confusion. Id.

In Sosnowski v. Wright Medical Technology, Inc., No. 11 C 59, 2012 WL 1030485, at *3 (N.D. Ill. Mar. 27, 2012), the court held that 510(k) clearance demonstrated compliance with industry standards and so supported summary judgment for the manufacturer on a design defect claim, even though the plaintiff criticized the process.

In Marcum v. DePuy Orthopedics, Inc., No. 1:12-CV-834, 2013 WL 1867010, at *6-7 (S.D. Ohio. May 2, 2013), the court held that clearance of a Class II hip device was relevant to whether punitive damages should be awarded under Ohio law, Ohio Rev. Code §2307.80(C), which barred punitive damages where FDA “approval or license” had been obtained.

In Booker v. C.R. Bard, Inc. (In re: Bard IVC Filters Products Liability Litigation), 289 F. Supp. 3d 1045, 1047-50 (D. Ariz. 2018), even though the district court mistakenly thought that clearance was not a safety determination, it ruled that the jury could be told how Bard complied with the 510(k) clearance process and said any potential confusion could be cured, if necessary, by a limiting instruction. It added:

It is worth noting that the absence of any evidence regarding the 510(k) process would run the risk of confusing the jury as well. Many of the relevant events in this case occurred in the context of FDA 510(k) review, and much of the evidence is best understood in that context. Attempting to remove any references to the FDA from the trial would risk creating a misleading, incomplete and confusing picture for the jury.

Id. at 1049.

For plaintiffs:

Courts have also allowed plaintiffs to introduce FDA evidence, at least where the FDA’s basis for refusing clearance was an express safety concern. For example, in Reeves v. AcroMed Corp., 44 F.3d 300, 308 (5th Cir. 1995),[12] the FDA’s concern that screws could break was admissible to show that the device was unreasonably dangerous per se under Louisiana law.[13]

The pain pump cases are particularly instructive because pain pumps are Class II devices that were 510(k) cleared for general surgical use but were not cleared for use in shoulder joint surgery. When the FDA refused to clear one manufacturer’s device and gave a lack of safety and effectiveness data as a reason, some courts held that this was admissible as evidence of a defect. For example, in Miller v. Stryker Instruments, No. CV 09-813-PHX-SRB, 2012 WL 1718825, at *2-3 (D. Ariz. Mar. 29, 2012), an FDA declaration that the manufacturer had failed to show the safety and effectiveness of the device was held relevant to a claim that the manufacturer should have done further testing.[14]

Issue of law

Typically, these opinions allow evidence to be introduced with an instruction to the jury concerning its use. However, the U.S. Supreme Court in Merck Sharp & Dohme Corp. v. Albrecht, 139 S. Ct. 1668, 1679-80 (2019) held that it was for the court, not the jury, to interpret the meaning of FDA action. If so, the proper way to put FDA action before the jury would be a jury instruction. In Batty v. Zimmer, Inc. (In re Zimmer Nexgen Knee Implant Products Liability Litigation), No. 11 C 5468, 2015 WL 5145546 (N.D. Ill. Aug. 31, 2015), the court perhaps anticipated this result. There, the court held that any relevant FDA information should be introduced by stipulation and reserved ruling as to whether other FDA evidence could be used if punitive damages came into consideration: “[T]o the extent this information is otherwise admissible, the court can inform the jury what a 510(k) clearance does and does not signify.” Id. at *14.

Sauce for goose but not gander?

An important limit on the use of FDA action by plaintiffs lies in the way in which FDA regulatory standards differ from those of the common law. In the cases discussed above, the FDA action held to be admissible was a specific finding. They would not be applicable to all FDA actions. That is because the FDA standard for action is more consumer-friendly than the common law standard. The FDA can act based on a device’s mere “association” with adverse events, without any proof that the device actually caused the events.

For example, device manufacturers are required to submit a Medical Device Report (MDR) when they “receive or otherwise become aware of information, from any source, that reasonably suggests” that their device “[m]ay have caused or contributed to a death or serious injury,” among other circumstances. 21 CFR § 803.50 (emphasis added). The trigger for reporting such an adverse event is substantially lower than a plaintiff’s burden of proving causation when making claims against a manufacturer for damages.

For this reason, courts have excluded from evidence an expert who relied on an FDA withdrawal letter that made no finding of causation. Rider v. Sandoz Pharmaceuticals Corp., 295 F. 3d 1194, 1201 (5th Cir. 2002). Similarly, another court has refused to allow causation testimony to be based on FDA labeling requirements where the FDA “uses a different standard than a court does to evaluate evidence of causation in a products liability action.” In re: Mirena IUS Levonorgestrel-Related Products Liability Litigation (No. II), 387 F. Supp. 3d 323, 356 (S.D.N.Y. 2019).

Reasons for excluding 510(k) evidence

The only opinion excluding evidence of 510(k) clearance before the ruling of the pelvic mesh MDL court was Zemaitatis v. Innovasive Devices, Inc., 90 F. Supp. 2d 631, 633-34 (E.D. Pa. 2000). The court denied a new trial. In doing so, it defended its decision to preclude the defendant’s expert from relying on FDA data. It said the defendant had not provided any authority to support admission of the evidence and that any error would be harmless. The court nevertheless added that 510(k) clearance was not “de novo review” and that the jury might be misled into thinking the FDA had conducted substantial testing. The court appears to have assumed that the clearance process at issue was the same as in Lohr.

Courts have also refused to allow plaintiffs to argue that the failure to gain 510(k) permission for off-label use is evidence that a device is defective for that use. For example, clearance might have been denied simply because there is no predicate device and not because the FDA believed the use was unsafe.[15] These cases have generated quotable phrases to the effect that the denial of clearance is not “about safety,” but they too have no relevance to a case where a device has in fact been cleared.[16]

From this survey, it seems plain that the Fourth Circuit was wrong when it said in Cisson that “the clear weight of persuasive and controlling authority favors a finding that the 510(k) procedure is of little or no evidentiary value.” 810 F.3d at 920. To be sure, there are a great many cases that pick up on Lohr’s dictum about 510(k)’s focus being on “equivalence, not safety,” 518 U.S. at 493, but they are typically preemption cases, not cases about the admission of evidence. Two of the five cases the Fourth Circuit cited to support its claim were preemption cases, not evidence cases.[17] Another concerned the interplay between 510(k) and HEW regulations.[18]

While two of the cases cited in Cisson did exclude 510(k) evidence, they, as discussed above, excluded evidence of the failure to get 510(k) clearance, not the grant of 510(k) clearance.[19] The FDA’s refusal of 510(k) clearance is ambiguous and so its relevance is doubtful. Given that the FDA’s standards are more protective of the consumer than the common law, admission of the evidence could easily mislead the jury.

On the other hand, the FDA’s grant of clearance under 510(k), in the normal case where the device falls within Class II, is not ambiguous. It is, as a matter of law, a finding that there is “reasonable assurance” of the safety of the device and that it does not present a “potential unreasonable risk of illness or injury” that would require it to be in Class III. 21 U.S.C. § 360c(a)(1)(B), (C). In that situation, the “great weight” of authority—outside of the pelvic mesh cases—is in favor of admitting the evidence.

[1] John Q. Lewis & Sarah Bunce, 510(k) Medical Device Trials Effective Use of FDA Evidence, 56 No. 9 DRI for Def. 40 (Sept. 2014).

[2] Jeffrey K. Shapiro, Substantial Equivalence Premarket Review: The Right Approach for Most Medical Devices, 69 Food & Drug L.J. 365 (2014) (citing Institute of Medicine (IOM), Medical Devices and the Public’s Health: The FDA 510(k) Clearance Process at 35 Years 4 (2011)).

[3] See Luther Munford, Guest Post – 510(k) Clearance of a Class II Device Can Be, as a Matter of Law, a Finding of No Potential Unreasonable Risk, Drug and Device Law Blog (March 1, 2019), https://www.druganddevicelawblog.com/2019/03/guest-post-510k-clearance-of-a-class-ii-device-can-be-as-a-matter-of-law-a-finding-of-no-potential-unreasonable-risk.html.

[4] 21 U.S.C. §360c(a), (e)-(i).

[5] U.S. Department of Health and Human Services, Food and Drug Administration, Center for Devices and Radiological Health, Center for Biologics Evaluation and Research, The 510(k) Program: Evaluating Substantial Equivalence in Premarket Notifications [510(k)], Guidance for Industry and Food and Drug Administration Staff, U.S. Food & Drug Administration, Regulatory Information, FDA Guidance Documents, 6-7 (July 28, 2014, current as of February 5, 2018), https://www.fda.gov/regulatory-information/search-fda-guidance-documents/510k-program-evaluating-substantial-equivalence-premarket-notifications-510k.

[6] 21 C.F.R. §807.93(a)(1).

[7] 21 CFR §807.92.

[8] Lewis & Bunce, supra note 1.

[9] Shapiro, supra note 2, at 389-91. The study also considered 510(k) exempt devices and counterfeit devices that were not reviewed by the FDA, and those categories comprised the remaining percentage of serious recalls.

[10] See, e.g., U.S. Government Accountability Office, Medical Devices: FDA Should Take Steps to Ensure that High-Risk Device Types are Approved through the Most Stringent Premarket Review Process 7 (January 2009), http://www.gao.gov/assets/290/284882.pdf (“We are recommending that FDA expeditiously take steps to issue regulations for each class III device type currently allowed to enter the market through the 510(k) process, including (1) reclassifying each device type into a lower class or requiring it to remain in class III and (2) for those device types remaining in class III, requiring approval for marketing through the PMA process.”).

[11] Shapiro, supra note 2, at 373-77.

[12] Any holdings of the Reeves court with respect to preemption must be read in light of Lohr, 518 U.S. 470 (1996).

[13] See also Corrigan v. Methodist Hosp., 874 F. Supp. 657 (E.D. Pa. 1995) (failure to clear screws for spinal use admissible in malpractice case).

[14] See also Placencia v. I-Flow Corp., No. CV10-2520 PHX DGC, 2012 WL 5877624, at *5-6 (D. Ariz. Nov. 20, 2012) (same); Pritchett v. I-Flow Corp., No. 09-CV-02433, 2012 WL 1340384, at *3-4 (D. Colo. Apr. 18, 2012) (same); Block v. Woo Young Medical Co. Ltd., 937 F. Supp. 2d 1028, 1046-47(D. Minn. 2013) (rejection by FDA relevant to failure to test pumps).

[15] See Rodriguez v. Stryker Corp., 680 F.3d 568, 574 (6th Cir. 2012); Healey v. I-Flow, 853 F. Supp. 2d 868, 877 (D. Minn. 2012).

[16] See also Todd v. Stryker Corp., No. 2:09-CV-01509-JAM, 2012 WL 2922727, at *5 (E.D. Cal. May 1, 2012); Pavelko v. Breg, Inc., No. 09-CV-01461-PAB-KMT, 2011 WL 782664, at *6 (D. Colo. Feb. 28, 2011); Forslund v. Stryker Corp., No. CIV. 09-2134 JRT/JJK, 2010 WL 3905854, at *4 n.5 (D. Minn. Sept. 30, 2010).

[17] Riegel v. Medtronic, 552 U.S. 312, 322 (2008) (approval of Class III device triggers express preemption); Duvall v. Bristol-Myers-Squibb, 103 F.3d 324, 329 (4th Cir. 1996) (denying preemption because §510(k) clearance based on pre-1976 device imposed no requirements).

[18] Almy v. Sebelius, 679 F.3d 297, 308 (4th Cir. 2012) (not arbitrary and capricious for HHS to require proof of safety and effectiveness beyond 510(k) clearance in order to justify Medicare reimbursement).

[19] Rodriguez v. Stryker Corp., 680 F.3d 568, 574 (6th Cir. 2012) (FDA denial of §510(k) clearance for lack of a predicate was not a statement that device was unsafe); Talley v. Danek, 179 F.3d 154, 160 (4th Cir. 1999) (because off-label uses are not illegal, failure to get approval for a Class III use was not negligence per se).