510(k) Clearance

Subscribe to 510(k) Clearance

It is now 2019, but we are still finding bits of leftover 2018 business on our desk and in our emails. Towards the end of last year, we encountered an avalanche of good rulings from the Southern District of Indiana in the Cook IVC filters litigation. Here is one we found hidden in the toe of our Christmas stocking: In re Cook Medical, Inc., IVC Filters Marketing, Sales Practices and Product Liability Litigation, 2018 WL 6617375 (S.D. Ind. Dec. 18, 2018). It is nice enough; it is not as if we are going to drive to the mall and return it. But there are some parts to it that we don’t love so much. Those parts are like ugly socks that we deposit in the bottom of a drawer and hope never to see again.

The plaintiff moved in limine to preclude evidence of 510(k) clearance of the medical device. The primary basis for such preclusion was Federal Rule of Evidence 402 – lack of relevance because the 510(k) process does not provide a reasonable assurance of the device’s safety and efficacy. That’s the argument, anyway. Prior to the pelvic mesh litigation, that argument was a sure loser. But, sadly, a couple of the pelvic mesh courts have swallowed this bogus argument hook, line and stinker. (Then again, we know of at least one recent non-mesh decision that rejected the no-510(k) argument, and we were so pleased that we deemed that decision one of the ten best of last year.)

As we have shown in a previous walk-through of this issue, the exclusion of 510(k) clearance is based on an over- or misreading of the SCOTUS Lohr decision, where the High Court contrasted the less rigorous 510(k) process with the Pre-Market Approval process. Lohr included loose language about how 510(k) clearance was limited to substantial equivalence with a predicate, rather than an independent demonstration of safety and efficacy. But SCOTUS itself subsequently reeled in that loose language in Buckman, recognizing that substantial equivalence was, in fact, a way of establishing safety. Moreover, the FDA itself subsequently tinkered with the 510(k) process and its characterization of it so as to make clear that 510(k) clearance is about safety. It is not as if the FDA would clear products it does not believe are safe. So where are we, or where should we be, when it comes to 510(k) clearance? Such clearance might not be enough to preempt a state law tort, but it is still relevant to legitimate defenses and should, therefore, be admissible.

Where does the S.D. Indiana Cook decision fit on the spectrum? It is probably more to the good (pro-defense) side, but not quite as far as we would like. The fact of FDA 510(k) clearance comes in. That’s good. At least the jury will not be under the misimpression that the company unleashed a product on the populace willy-nilly, with no governmental oversight. But … well there’s a big but. (We knew a fellow defense hack who never missed an opportunity to use that phrase in diet drug litigation to get a cheap laugh).

Interestingly, the S.D. Indiana analysis turned on an interpretation of Georgia’s risk utility test for design defect cases, and Georgia law was also at issue in one of the very bad mesh decisions in this area (that we will not and, for reasons of our existing litigation entanglements, cannot name.) Georgia law incorporates the concept of ‘reasonableness,’ i.e., whether the manufacturer acted reasonably in choosing a particular product design. FDA clearance is relevant to such reasonableness. The S.D. Indiana court held that both the plaintiff and the defendant, through appropriate expert testimony, “will be permitted to tell the jury about the role of the FDA in its oversight of medical device manufacturers, the regulatory clearance process for devices like IVC filters, and [Cook’s] participation in the 510(k) process and its compliance (or lack thereof) with the process.”

What do we think of this ruling? It is both good and bad. It is good (and absolutely correct) that clearance comes in. But it is bad, because it seems to welcome plaintiff ‘expert’ testimony that instructs the jury on the law. Why should an expert be able to tell the jury that the company did not comply with the FDA process? Isn’t the fact of clearance itself proof that there was compliance? Is the plaintiff arguing that the FDA erred when it cleared product? Shouldn’t that be up to the FDA? Given the extensive publicity accompanying any mass tort litigation, wouldn’t the FDA have corrected its error, if there really was error? Or is the plaintiff arguing that the FDA cleared the product only because the company hid data and hoodwinked the FDA? If we are not squarely in Buckman-land, are we not at least Buckman-adjacent? Don’t the selfsame policies of deferring to the FDA apply? In other words, isn’t plaintiff’s anti-clearance position preempted?

Thus there is a part of the S.D. Indiana’s ruling that reeks to our (admittedly oversensitive) noses. The defendant is not permitted to present evidence or argument that the FDA’s 510(k) clearance of the device constitutes a finding by the FDA that the device is safe and effective. As set forth above, that ruling is factually incorrect. By contrast, the plaintiff “may present evidence that the FDA clearance process only requires substantial equivalence to a predicate device, that 501(k) regulations are not safety regulations, that Plaintiff’s filter placement was “off label,” and the like.” Wrong, wrong, and whatever “the like” means, we’re sure that’s wrong, too. The 510(k) process does, indeed, address safety. The notion is that relying on a predicate device that was already approved or cleared is a good proxy for safety. There is plenty of regulatory history showing that by devising the 510(k) process, the FDA was not waving bye-bye to the value of safety. Meanwhile, we will all be treated to a blow-hard plaintiff regulatory expert who will take us on a tour of FDA regulations and company documents to tell a tale of a bad company and an overmatched federal government. Some fun. This expert testimony amounts to a preview of the plaintiff’s closing argument. It is so gruesome that we have heard some defense lawyers say that they would just as soon omit the regulatory story altogether. But that’s defeatism. The right result would be for the fact of 510(k) clearance to come in, and then full-stop. There is no need for expert interpretation or interpolation. So for all the trial judges out there who complain that drug and device trials go on too long, here’s an answer: shut down the expert testimony that purports to teach the jury about the regulatory process “and the like.” Always beware of “the like.” And be on guard against “etc.” and “whatnot” as well.

But let’s get back to the good bits of the S.D. Indiana Cook decision. The defendant will be allowed to offer evidence that the device was never recalled by the FDA, that the FDA never observed any violation of the company’s quality system during its inspection between 2000 and 2014, and that the FDA never took any enforcement action against the company. This evidence is relevant to the defense that the design and development decisions were reasonable and that the product is safe. At least this court had some sense of balance.

The S.D. Indiana decision also dealt with the defendant’s effort to use an expert biomedical engineer to talk about the low rate of complaints. She worked at the FDA for over twenty years in various positions and currently serves as a consultant to companies seeking to obtain FDA approval or clearance of medical devices. The expert cited evidence that the device had a fracture rate of 0.066% and perforation rate of 0.153%. The defense expert calculated the occurrence rate by dividing the total number of complaints received by the company (numerator) by the product’s total sales (the denominator).

The plaintiff challenged this expert testimony because “there is no way to know whether these numbers are accurate. Some patients/hospitals may not report adverse events, and some IVC filters may have been sold to hospitals but not used in patients.” The company responded that it did not intend to offer the complaint/occurrence rates as the actual complication rates for the product. With that limitation, the Cook court held that the defense expert’s testimony was admissible.

Here in Philadelphia, less than a week before the first day of Spring, it is sunny and calm, albeit a bit chilly. Not so last week, when we were hit with the second Nor’easter in less than a week.  As much as sixteen inches of snow fell in some suburbs, and thousands of people, some of whom had just regained electric power after the last storm, were knocked into the dark and cold when tree limbs felled by heavy snow took power lines with them.  For the second time in as many storms, we found ourselves in a pocket on the low end of predicted accumulations.  We only got about six inches of snow, and it started melting quickly when temperatures rose in the evening.  And, again, our end of the street did not lose power.  (Assuming cooperation from the technology gods, this no longer affects us, as we happily parted with some retirement savings a few years ago in favor of a whole-house generator, after a four-day outage during which we tested long-held friendships by descending with the Drug and Device Law Menagerie in tow.)  The only “blip” was a scary near-miss.  We were out walking the Drug and Device Law Little Rescue Dogs at the height of the storm when a huge tree limb crashed down on a spot where we had stood ten seconds earlier.  Reminds us not to sweat the small stuff – this could have been tragic.

But save for that significant footnote, it was all pretty good.   And our assessment of today’s decision is similar.  In In re Bard IVC Filters Prods. Liab. Litig., 2018 WL 1109554 (D. Ariz. Mar. 1, 2018), the court ruled on a number of motions in limine in advance of a mid-March trial.  The plaintiff alleged that she was injured when defendant’s filter, inserted in her inferior vena cava (“IVC”) to capture blood clots, migrated and fractured.  The filter that was implanted in the plaintiff, a Class II device, was granted 510(k) clearance by the FDA because of its “substantial equivalence” to an earlier filter model the defendant had marketed.  And that is the source of the most interesting rulings.

Defendants’ Motion to Exclude Evidence of Complications Associated with Predicate Device

Earlier, the plaintiff had moved to exclude evidence of the 510(k) clearance process, a motion that the defendants “vigorously opposed.” In re Bard IVC Filters, 2018 WL 1109554 at *1.  The court agreed with the defendants, “noting that the FDA grants 510(k) clearance only where the device is as safe and effective as a predicate device and does not raise different questions of safety and efficacy than the predicate device.” Id. (internal punctuation and citations omitted).  As such, compliance with the 510(K) clearance process was “certainly probative under Georgia law on the issues of reasonableness of the design, manufacture, and warnings of the [cleared filter]” as well as issues related to the plaintiff’s punitive damages claim. Id. Georgia law, hmmmm…. The Fourth Circuit’s Cisson misstep, going 180° the other way, was also Georgia law, as we discussed here.  Needless to say, we like this one better.

But having fought for admission of evidence of the 510(k) clearance process, the defendants were faced with the plaintiff’s intention to introduce evidence of complications with the predicate device. The defendant moved to exclude this evidence, arguing that the complications were not “substantially similar” to the issues the plaintiff experienced with her device.  The defendant argued that, ‘[b]ased on clinical experience with the [predicate] filter, [it] made several significant changes to the [new] filter.” Id. Because the plaintiff received the new filter, which had been “changed significantly” from the predicate device, the defendants argued that problems with the predicate device  had no relevance to the plaintiff’s case.

The court denied the motion, noting that the defendant had “avowed to the FDA that the design, material, components, fundamental technology, . . . and intended use featured with the [new filter] are substantially equivalent to those featured with the predecessor [filter].” Id. The defendants further asserted that “one of [the] goals in developing the [new] filter was to reduce the number of incidents of filter fracture and migration that [it] had observed with the [predecessor].” Id. The court concluded, “Given these facts, . . . [the defendant’s] knowledge of problems with the [predecessor filter] is relevant to central issues in this case” – whether the defendant properly designed the new filter to correct the problems, whether it failed to properly warn about the problems, and whether its conduct justified an award of punitive damages. Id. at *2.

This ruling presents a quandary. We have spent many years defending manufacturers of Class 2 medical devices and advocating for admission of evidence of 510(k) clearance. We can appreciate the tension and the difficulty of drawing appropriate evidentiary lines.  But evidentiary lines and regulatory requirements are different.  This is a variant of the same fallacy underlying the 510(k) admissibility point discussed above.  FDA regulatory requirements and admission of evidence in civil litigation are two different things, created for different purposes, and often don’t mix well.

That is true here. For one thing, there is more to 510(k) than just substantial equivalence (“SE”).  The entire relevant part of the FDCA allows for an SE finding where the device being considered:

(ii)(I) has different technological characteristics and the information submitted that the device is substantially equivalent to the predicate device contains information, including appropriate clinical or scientific data if deemed necessary by [FDA] that demonstrates that the device is as safe and effective as a legally marketed device, and (II) does not raise different questions of safety and effectiveness than the predicate device.

FDCA §513(i)(1)(A) (emphasis added), So a substantial equivalence finding by the FDA does not mean that the predicate device is technologically the same – or necessarily even close – as long as it “does not different questions of safety and effectiveness.”  That’s a lot different from the “substantially similar” standard generally used for admission of other occurrence evidence.

Admission of evidence of 510(k) clearance should not automatically open the floodgates to evidence of problems a plaintiff didn’t’ experience with a device she didn’t receive. The other occurrence could well arise from a “different technological characteristic.”  That would require a deeper dive into the regulatory pedigree of the device than occurred here.  We are thinking about this issue for a future post. Meanwhile, here are some highlights of the court’s less complicated rulings:

Defendants’ Motion to Exclude FDA Warning Letter

The defendants moved to exclude an FDA warning letter because: 1) it was an “informal advisory statement” by FDA; 2) it was issued more than seven years after the plaintiff received her IVC filter; and 3) the specific topics in it were not related to the filter the plaintiff received or to any issue in the case. Id. at *3.  The plaintiff countered that the letter was evidence to counter the defendant’s assertion that the FDA had never expressed concern about the defendant’s filters or taken any action against the defendant.   The court held that several sections of the warning letter lacked probative value because they addressed topics that were not at issue in the litigation.   Others addressed a generation of filter that was developed after the filter the plaintiff received.  The court granted the motion in limine with respect to those topics, but reserved for trial the question of the admissibility of topics related to the defendant’s handling and reporting of filter failures.  The court did conclude that, if the letter should become relevant at trial, it would be admissible under the public records exception to the hearsay rule.  Whatever happens later, the defendant’s first point about warning letters being “informal” and “advisory” was spot on.  We examined that issue in great detail here and here.

Plaintiffs’ Motion to Exclude References to IVC Filters as “Lifesaving” Devices.

The plaintiff sought to preclude the defendants from “putting on a ‘filters save lives’ defense” or from describing the filters as “lifesaving” or “life-extending.” Id. at *4.  The court denied the motion, holding that, under Georgia’s risk-utility analysis, “evidence concerning the benefits of IVC filters is directly relevant. . . .”

Plaintiffs’ Motion to Exclude “Standard of Care” Evidence

The plaintiffs sought to exclude the defendants from presenting evidence that IVC filters are “within the standard of care for medical treatment of pulmonary embolisms.” Id. at 5 (citations omitted).   The court denied the motion, agreeing with the defendants that “evidence regarding the use and benefits of IVC filters, and when they are called for, will be relevant to the jury’s risk-utility analysis, as well as evaluation of the failure to warn claims and [the surgeon’s] decision to implant” the defendant’s filter in the plaintiff. Id. In response to the plaintiffs’ concern that the jury could be confused about the standard of care to apply, the court conceded that “the standard of care for . . . design and marketing of IVC filters is entirely different from the medical standard of care for when filters should be used to treat patients.” Id. As such, the court held, the parties should refer to the “medical standard of care” when referring to the standard for implanting the filters, and should seek a clarifying instruction if warranted.

Plaintiffs’ Motions: “Nonparty at Fault” and “Intervening Cause”

Under Georgia law, a defendant may identify a “non-party at fault” and present evidence of that party’s fault to the jury. The jury may then include the non-party in its apportionment of fault and damages.  In this case, the defendant identified a diagnostic radiologist as the only “nonparty at fault,” alleging that the radiologist’s failure to report the condition of the filter to the plaintiff’s physicians was a cause of the plaintiff’s injuries.  While this was the only non-party whose fault the defendants intended to prove to the jury, the defendants argued that they should be permitted to introduce other doctors’ conduct, under a separate legal doctrine, as “intervening causes” of the plaintiff’s injuries, breaking the chain of causation between the defendant and the plaintiff and relieving the defendant of liability under Georgia law.  The plaintiffs moved to exclude evidence that  doctors not identified as “nonparties at fault” were “intervening causes” of the plaintiff’s injuries.  The court denied the motion, holding “. . . [A]lthough Defendants will be precluded from arguing that [other doctors] were negligent or at fault for purposes of apportioning liability, the Court cannot say that they should be precluded from asserting the fault of these nonparties as an intervening cause” of the plaintiff’s injuries.  The court emphasized that it would be careful to instruct the jury about the distinction between the two doctrines.

Plaintiffs’ Motion to Exclude Statements from Professional Associations

The plaintiffs sought to exclude evidence of statements of professional associations, trade groups, societies of physicians, and the like, arguing that the statements were hearsay, and, introduced through non-experts, would “evade Daubert scrutiny and would violate Rule 403.”  The court reserved its rulings for trial, holding that it could not grant the motion because the plaintiffs did not identify particular statements.  But it held that, on the present record, it could not conclude that the probative value of certain statements was outweighed by the potential for prejudice or that testimony about the absence of certain statements constituted inadmissible hearsay.  We see an analogy to how the federal rules treat learned treatises for hearsay purposes.

All in all, a pretty good day, with the exception of the quandary created by the rulings related to the admissibility of 510(k) evidence. That one isn’t as black and white as the ruling that was made.  We will keep you posted.