Photo of Lisa Baird

In this, the Year of Our Lord 2025, that title could apply to so, so, so many things.  Soooooo many things.

But this is the Drug and Device Law Blog, product liability is our niche, and what we are referring to is the new, 2024 European Union Product Liability Directive (the “PLD”), also known as Directive (EU) 2024/2853 of the European Parliament and of the Council of 23 October 2024 on liability for defective products and repealing Council Directive 85/374/EEC (Nov. 18, 2024).

We have spoken and written about the issue before.  And now we are starting to jump up and down to get attention on this.

Why the cause for alarm?  Look at the litigation funding industry.  Rumor has it that they are so excited about the new PLD, they already have set up camp in Amsterdam and London in anticipation of the disputes to come.

You may be thinking to yourself, I’ve defended mass torts in some of the most plaintiff-friendly jurisdictions in the U.S., how bad can it be?

How about not actually requiring proof of defect, particularly because the product at issue is a life-saving medical device:

Some products, such as life-sustaining medical devices, entail an especially high risk of causing damage to people and therefore give rise to particularly high safety expectations. In order to take such expectations into account, it should be possible for a court to find that a product is defective without establishing its actual defectiveness, where it belongs to the same production series as a product already proven to be defective.

PLD, at 19 (emphasis added). 

How about holding subsequent remedial measures against manufacturers?  The silver lining (?) is that subsequent remedial measure “should not in itself lead to the conclusion that a product is defective.”  PLD, at 22, id. at 60. 

How about a rebuttable presumption of defectiveness where a defendant has failed to comply with an “obligation to disclose information.”  PLD, at 31, id. at 67 (“relevant evidence”). 

How about a rebuttable presumption of defectiveness “in a case of obvious malfunction… since it is unnecessarily burdensome to require a claimant to prove defectiveness when the circumstances are such that its existence is undisputed”  PLD, at 31, id.at 67.  Obvious to whom?  Undisputed by whom?   

How about a rebuttable presumption of causation “[w]here it has been established that a product is defective” (maybe through one of the defectiveness presumptions above) “and the kind of damage that occurred is, based primarily on similar cases, typically caused by the defectiveness in question” because “the claimant should not be required to prove the causal link and its existence should be presumed.”  PLD, at 32, 68.  Should we call this the “you win one, you win them all rule” for plaintiffs?

Perhaps worst of all is the rebuttable presumption of both defectiveness and causation, which will arises just because the issues are hard:

National courts should presume the defectiveness of a product or the causal link between the damage and the defectiveness, or both, where, notwithstanding the defendant’s disclosure of information, it would be excessively difficult for the claimant, in particular due to the technical or scientific complexity of the case, to prove the defectiveness or the causal link, or both…. While a claimant should provide arguments to demonstrate excessive difficulties, proof of such difficulties should not be required.

PLD, at 33 (emphasis added); see also PLD, at 68. 

And again, our clients’ essential products are called out as deserving of punishment through this evidentiary shortcut designed to lead straight to a liability finding:

Technical or scientific complexity should be determined by national courts on a case-by-case basis, taking into account various factors. Those factors should include the complex nature of the product, such as an innovative medical device.

PLD, at 33 (emphasis added). 

Also mind-boggling are the rationales given for the new PLD.  For example:

Liability without fault on the part of economic operators remains the sole means of adequately addressing the problem of fair apportionment of risk inherent in modern technological production. 

PLD, at 2 (emphasis added).  Really?  How about the regulatory systems in place in the E.U. already?  They work quite well for the pharmaceutical and medical device sectors.  How about national health care programs that provide medical care to all without risk of personal bankruptcy or the inefficiency of litigation, and the already-existing possibility of recovery from pharmaceutical manufacturers, when necessary and if appropriate, by those national health care programs?

One very modest (very modest) beneficial (or at least not openly harmful) provision that we see in the new PLD is a sort-of regulatory compliance defense: 

An economic operator . . .  shall not be liable for damage caused by a defective product if that economic operator proves . . . that the defectiveness that caused the damage is due to compliance of the product with legal requirements.

PLD, at 69. 

Another is the “state of the art” defense:

An economic operator . . .  shall not be liable for damage caused by a defective product if that economic operator proves . . . the objective state of scientific and technical knowledge at the time the product was placed on the market or put into service or during the period in which the product was within the manufacturer’s control was not such that the defectiveness could be discovered.

PLD, at 70.  But don’t get too excited:  individual E.U. nations can opt out of this one.  PLD, at 79-80. 

There is a lot more about the new PLD that warrants discussion. 

For example, it isn’t only pharmaceutical and medical device manufacturers who are in the cross-hairs.  If there is a nuclear accident, or you make free and open-source software, you are in the clear.  PLD, at 4, 47.  But other software, AI, digital manufacturing files (as for 3D printers), electricity, and raw materials are products falling within the PLD.  See PLD, at 8, 49-50.  And the new PLD applies to goods and related services.  See PLD, at 11.    

The new PLD was published in the Official Journal of the European Union on November 18, 2024.
EU Member States now have two years to “transpose” it into their national laws, i.e. by December 9, 2026, and the new PLD will apply to products put on the market in the EU after December 9, 2026.

In the coming months, we will continue to pick apart the new PLD—and continue to sound the alarm.

Photo of Lisa Baird

This post is from the non-Butler Snow side of the blog.

When you represent medical device manufacturers in product liability litigation, you will deal with allegations that a device broke or failed because of what it was made from, and you will encounter both experts and “experts” (scare quotes intended) in materials science. 

Materials science is the interdisciplinary study and analysis of product composition with application to the design, development and manufacturing of real products.  It is something we have found weirdly interesting, ever since an amazing materials scientist explained that a favorite blazer with a nubbly texture never wrinkled because it had “predetermined collapse points.”  Because of those predetermined collapse points, the garment never took on additional collapse points (aka wrinkles), no matter how badly it was squished in a suitcase.

3D printing/additive manufacturing is another area of interest to the blog, and another area where materials science matters quite significantly.  In talking with another great materials scientist, Ming Tang, we learned that the simple act of changing the orientation of how an object is 3D printed may change its properties in pretty significant ways.  Ming and his colleague Larry Eiselstein were even kind enough to send us an article, 3D-Printed Metallic Medical Devices: Increasing Interest from Medical Device Manufacturers, Opportunities, and Challenges, in which they explained this and other materials science issues related to 3D printing in the medical device space.  Even to a lay person, it is a quick and understandable read, not to mention eye-opening and interesting.

With that short detour over , turn with us now to Hill v. Medical Device Bus. Srvcs., Inc., No. 3:21-cv-0440, 2024 U.S. Dist. LEXIS 140272, 2024 WL 3696481 (M.D. Tenn Aug. 7, 2024), a case in which the plaintiff’s materials science expert was challenged and excluded, for good reason.

The plaintiff in Hill had two hip replacement surgeries.  A 2014 surgery was for the initial implant of his total hip replacement system, and a 2015 surgery was to replace certain components, although one component, the femoral stem, was left unchanged.  Five years later, the femoral stem component fractured, and the entire hip implant construct was explanted and replaced. 

According to the court, the parties seemingly agreed that the femoral stem component broke due to a small flaw in its metal, “but they disagree[d] as to how that flaw came to exist”:

Defendant’s experts assert that a flaw was introduced to the product during surgery from the use of electrocautery… [whereas] Plaintiffs assert that a flaw was introduced to the product during the manufacturing process.

Hill, 2024 U.S. Dist. LEXIS 140272 at *4.

As our readers know, expert testimony is measured against the standard in the updated Federal Rule of Evidence 702, which emphasizes the importance of the gatekeeping function of our courts:

A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if the proponent demonstrates to the court that it is more likely than not that:

(a) the expert’s scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue;

(b) the testimony is based on sufficient facts or data;

(c) the testimony is the product of reliable principles and methods; and

(d) the expert’s opinion reflects a reliable application of the principles and methods to the facts of the case.

To its credit, the Hill court recognized the changes to Rule 702 were meant to give it more teeth, and that courts “can exclude a conclusion if it is based on methods that are unreliable (and thus do not serve to reliably substantiate the conclusion[)]”.

The Hill court also realized that Sixth Circuit cases applying the pre-amendment version of Rule 702 might be amongst the “misguided cases” that led to rules Advisory Committee to update Rule 702.  But—in the only really confounding part of the opinion, footnote 9—the court believed that it had to follow Sixth Circuit authorities interpreting the previous version of Rule 702 even though Rule 702 itself has changed.  See here for contrary Supreme Court precedent recognizing that rules changes, like statutory changes, overturn prior judicial precedent.

The good news is, the court did not need to resolve any finer points of the pre- versus post-Rule 702 amendment test to conclude that the plaintiffs’ materials science experts’ opinions were inadmissible.

Plaintiffs’ first expert was a well-credentialed materials scientist who opined that the critical flaw in the implant’s femoral stem was introduced during manufacturing, and that she had ruled out all other potential causes for the flaw.  But, as the court recognized, she did not actually have any evidentiary basis for ruling out those other potential causes: 

For example, she ruled out that the failure was caused by the Implant being placed at the improper angle during surgery, but [Plaintiffs fail] to explain how she can give that opinion when (as Plaintiffs do not dispute) she does not know the proper angle at which the Implant should have been placed.  If she does not know what angle is improper, then she lacks a basis for saying that an improper angle did not cause the break.  Similarly, she lacks a basis to rule out “trauma and blunt force” during surgery . . . because she did not know anything about how the surgery was conducted.

Hill, 2024 U.S. Dist. LEXIS 140272 at *21.

Moreover, although this expert testified it was possible for the flaw to have been introduced during manufacturing, the expert did not examine whether the defendant had appropriate measures in place to control the size of flaws during manufacture of its metal implants.  Thus, she couldn’t actually tie the flaw in the plaintiff’s implant to the manufacturing process or to the reason the plaintiff’s implant broke.

With the plaintiffs’ first expert out, the second expert also had to be excluded:  part of his opinion merely parroted the first expert and thus was unreliable, and the rest of his opinion was predicated on statements that the first expert never expressed.

With both of plaintiffs’ experts out, the Court concluded (quite rightly) that the burden of proof meant it did not need to consider challenges to the defendant’s experts and instead could just move on to the defendant’s motion for summary judgment.  

Without experts to establish causation, summary judgment should have been—and was—a relatively easy grant.  But plaintiffs tried one last gambit, the res ipsa loquitur-like malfunction theory:  that the factfinder can infer defectiveness merely from the existence of an alleged malfunction and a negation of other causes. 

We don’t buy the malfunction theory, just as we look askance at other variations of res ipsa loquitur.  When the malfunction theory is accepted, it is supposed to be fairly rigorous:  If the plaintiff can show that the product malfunctioned, and if the plaintiff also can negate all other causes for the malfunction other than a product defect, then the malfunction theory is applicable and an inference of a product defect is permissible.

But we find it often is trotted out just because the plaintiffs’ counsel or experts haven’t done their homework and are missing key pieces of evidence.  The device must have had a defect, they will argue, because devices don’t just fail. 

But the reality is that medical devices, particularly implanted medical devices, do fail for unknown reasons—and plaintiff’s second expert admitted as much here.  And when the plaintiff has not actually negated all other causes of malfunction, then the malfunction theory just doesn’t apply:

[W]hen a device is known to fail for unknown reasons, it is pure speculation that a failure is attributable to a manufacturing defect and not some other unknown cause without evidence supporting one cause over another.

Hill, 2024 U.S. Dist. LEXIS 140272 at *49.

The plaintiff in Hill certainly did not negate all other potential causes of malfunction, and so the court came to the right conclusion:  The defense wins.

An appeal has been filed, so we shall see if the Sixth Circuit agrees.

Our thanks to Robyn Maguire and Sarah Jin of Barnes & Thornburg for sending the case and a congratulations as well for the nice win!

Photo of Stephen McConnell

In the wake of last week’s webinar that we DDL bloggers presented on the best and worst cases of 2024, we received many kind notes from clients and colleagues.   The webinar was apparently a painless and – dare we say it? – even fun way to earn 1.5 CLE credits.  But we also heard from some in-house lawyers that they do not all subsist on a diet of straight product liability. Company legal departments harbor lawyers who work on a variety of matters, including intellectual property, commercial disputes, and sundry other areas.  Their point was that drug and device in-house attorneys also like to hear about those other legal issues. 

We take their point.  Accordingly, today we will discuss Nexus Pharmaceuticals, LLC v. Long Grove Pharmaceuticals, LLC, 2025 U.S. Dist. LEXIS 6058, 2025 WL 81877 (D. Mass. Jan. 13, 2025), a lawsuit brought under the Lanham Act, 15 U.S.C. section 1125(a).  The claim was false advertising. There was no personal injury alleged in the case. Instead, the alleged injury was to sales.  It was a dispute between two pharmaceutical companies. (Truth be told, we are tempted to remain mute when we see such a dispute. We do not want to stumble into client conflicts or hostilities. But this case implicates interesting issues such as a misbegotten effort at private FDCA enforcement, so here we go.) The parties were competitors.  Both sold a drug product used to enable diagnostic imaging. Go to the opinion (which is quite short) if you want to learn the details, but the key chronology in the case goes like this:


April 2023 – the FDA identified a shortage of the drug product.


June 2023 – the defendant bought the NDA for the product and any remaining stock from a bankrupt company. Because of the drug shortage, the FDA permitted the defendant to sell the old stock. In accordance with an FDA guidance, the defendant sent out a Dear Healthcare Provider letter that it would distribute the drug product to “address this critical drug shortage.”  That same language showed up on the defendant’s website. 


September 2023 – the plaintiff obtained FDA approval to market a generic version of the drug product. 


December 2023 – the FDA declared that the shortage was “resolved.”


But – and here’s the rub – even after the FDA’s declaration that the shortage was resolved, the defendant continued to distribute the drug product and continued to state on its website that the shortage still existed. The plaintiff asked the defendant to stop selling the product because the shortage had ended. The defendant refused, maintaining “that it had permission from the FDA, pursuant to its drug shortage authority, to continue selling” the product.  


Then the plaintiff filed this lawsuit, alleging that the defendant’s continuing statements about the drug shortage misled customers into believing that the plaintiff’s product was not available and that only the defendant could provide the product.  As a result, so the complaint alleges, the plaintiff lost customers to the defendant. The complaint also stated that the defendant was able to undercut on price because it was selling product from a bankrupt company that had stored the product in conditions unknown to the plaintiff or the public. 


The lawsuit was based on the Lanham Act and alleged unfair competition and false advertising. It sought a declaratory judgment against the defendant, plus actual damages. The complaint referenced one actual customer who had switched to the defendant’s product. 


The defendant moved to dismiss the complaint on several grounds, including that the plaintiff was essentially challenging the defendant’s compliance with FDA regulations, and that “the Lanham Act cannot supply a basis for relief from a statement that FDA expressly mandated a manufacturer to make.”  Those are interesting assertions and there’s a whiff of preemption, though displacement of one federal legal or regulatory regime by another does not really constitute preemption.  


But the Nexus court dodged these interesting assertions by deciding the much simpler issue of whether the claim was sound under the Lanham Act.  It was not, because 

the Lanham Act requires that the alleged misrepresentation be about the products at issue.  To prove a Lanham Act claim for unfair competition and false advertising, the plaintiff must demonstrate that:

1. The defendant made a false or misleading description of fact in a commercial advertisement about his own or another’s product;

2. The misrepresentation was material in that it likely influenced a purchase decision;

3. The misrepresentation actually received had the tendency to deceive;

4. The defendant placed the deceptive statement into interstate commerce; and

5. The plaintiff was likely injured by either a direct diversion of sales or loss of goodwill. 

Here, the plaintiff’s Lanham Act claim stumbled at the first hurdle.  The allegedly false claim that there was a shortage did “not refer to any inherent quality or characteristic” of either the plaintiff’s or defendant’s product.  Statements relating to the marketing method of a product are not related to product qualities or characteristics. (The Nexus court cited several other district court opinions that came out with similar holdings). The plaintiff did not allege that the defendant had made a false/misleading statement about the actual age of the product or its quality. It appears that such an allegation might have saved the Lanham Act claim, but we’ve already wandered far afield of our core knowledge, so we’ll leave that speculation to our readers who are smarter than we are either generally or merely with respect to the Lanham Act. 

Photo of Michelle Yeary

Today’s case — In re Tepezza Mkt’g, Sales Pracs., & Prod. Liab. Litig., MDL No. 3079, 2025 WL 81338 (N.D. Ill. Nov. Jan. 10, 2025) – doesn’t really break new ground on the substantive law, in fact it is a split decision.  But it comes from an unusual procedural posture worth pointing out. 

The Tepezza MDL was created in June 2023 and shortly thereafter the court ruled on a pending motion to dismiss in a single case.  The court denied the motion as to plaintiff’s pre-market design defect claims.  We were not particularly fond of the ruling and said so here.  It appears that since that time, the court decided to create a group of “bellwether discovery” cases.  The idea is that motions to dismiss get teed up and ruled on in the 12 “bellwether discovery” cases to establish the scope of discovery.  We can’t help but be intrigued with any process which has as a component setting boundaries on discovery.  Realistically, particularized rulings on plaintiffs’ claims could lead to less discovery.  For instance, certain time periods could be deemed out of bounds.  Or, if design defect claims are dismissed, that might limit the amount of research and development discovery that is warranted.  Furthermore, tethering the discovery to the claims of 12 actual plaintiffs, regardless of the rulings on the motions, could also be potentially helpful to defendants.  Often, discovery against the manufacturer in an MDL tosses open the doors to the company at large because plaintiffs claim they are taking discovery for all plaintiffs, for all purposes, for all claims, for all times.  Putting discovery in context, including the context of 12 individual plaintiffs, almost always inures to the benefit of the defense.  For our readers who spend a significant time slogging it out with plaintiffs on discovery disputes, we’ll keep our eyes on whether this discovery bellwether generates any useful precedent or tips going forward.

Now to turn to the substance.  Plaintiffs in the Tepezza MDL allege “hearing loss and/or tinnitus” from the use of an FDA-approved prescription biologic for thyroid eye disease.  Defendants challenged the bellwether plaintiffs’ failure to warn, design defect, and fraudulent misrepresentation.  They won one, lost one, and split one. 

On failure to warn, defendant argued that the claims were preempted because plaintiffs failed to adequately plead that defendant could have used the CBE process to change its label before September 2022—after plaintiffs had stopped their treatment.  First, we should point out that while Tepezza is a biologic, the CBE regulation is substantively identical to that for prescription drugs.  So, to avoid preemption, a plaintiff “must plead that the manufacturer had newly acquired information that showed a causal association between the drug and an effect that warranted a new or stronger warning, precaution, or adverse reaction in the label.”  In re Tepezza, at *4.  In that situation, the manufacturer does not need FDA approval to change its label; therefore, no preemption.  Here, plaintiffs cited to a clinical study that came out in September 2020 that showed a sufficient increase in the incidence rate to arguably satisfy the CBE requirements.  Id. at *5.  Defendant argued that the study could not form the basis for a CBE label change because it was not a “clinical trial,” but the court found “criticism of the quality of [the study] . goes to the sufficiency of the evidence proving the allegations, not the sufficiency of the allegations.  Since all of the bellwether plaintiffs used Tepezza after September 2020, their failure to warn claims were not preempted at the pleadings stage. 

On design defect, plaintiffs failed to meet the TwIqbal pleading standard.  In fact, plaintiff only made boilerplate allegations offering no facts and giving the court no basis from which to infer what aspect of the product makes it defective.  Plaintiffs tried to argue that under at least Florida and New York law, they didn’t need to allege a design defect before discovery because they do not have the technical information required to do so.  However, plaintiffs relied on cases in which the complaints contained many more factual details than they allege here.  Plaintiffs don’t have to prove a specific defect in their complaint, but they have “to do more than assert the conclusion that Tepezza is defective.”  Id. at *7.  Plaintiffs’ allegations about dosage were not supported by the evidence they pointed to in the complaint.  Moreover, simply alleging that dose is too high does not satisfy the need to allege an alternative design.  Id. at *9.  With design defect dismissed, plaintiffs asked for permission to amend their complaints—long after the MDL deadline to do so.  They claimed that they had new evidence that defendant was conducting a clinical trial on a new delivery method.  Not only was the study announced months before the pleadings deadline, but it was also not scheduled to be completed until 2026 and had no safety or efficacy results as of yet.  The ”mere initiation of a study” does not “plausibly suggest” a defect in the current delivery method or indicate that the method being studied will turn out to be safer or more efficacious.  Plaintiff did not offer good grounds for their motion to amend, and it was denied.

That left only plaintiffs’ fraudulent misrepresentation claim which was essentially the same as their failure to warn claim.  Plaintiffs’ fraud claim was based on two allegations—that defendant overstated the efficacy of the biologic and defendant misstated the incidence rate.  Applying Rule 9’s heightened pleading requirement for fraud, the court found both allegations were lacking.  The first was simply not supported by any factual allegations (“plaintiffs’ allegations regarding efficacy are highly generalized”).  Id. at *10.  As to the second, plaintiffs had conflicting allegations, including statements that the incidence rate on the label was accurate.  “So, Plaintiffs’ assertion of falsity is not well pled.”  Id. at *11. While the fraud itself was not sufficiently pleaded, the court found plaintiffs’ fraud damages need not be pleaded with particularity.  And defendants raised one more issue—that because the fraud claim is based on the FDA-approved label, it is actually a preempted fraud-on-the-FDA claim.  Here the court concluded it had insufficient information on which to base a decision. Because the fraud claim is substantively the same as the failure to warn claim, a ruling on Buckman preemption would not change the scope of discovery, and therefore the court deferred the issue.  So, on fraudulent misrepresentation the final ruling is defendant’s motion to dismiss is granted in part (fraud allegations), denied in part (damages); and continued in part.  Except as to the Pennsylvania plaintiff because Pennsylvania does not recognize fraud claims based on a failure to warn theory.  Her claim was dismissed with prejudice. 

Like we said, a 50/50 split decision on the claims—but a possible case to watch for discovery. 

Photo of Bexis

Back in 2008, we wrote a post, No, Bu shi, Non, Iie, Nada, Nyet…., collecting a significant body of law holding that overseas defendants are not required to translate, at their expense, documents prepared in their non-English home languages.  The other day we came across Sessoms v. Toyota Motor Sales, U.S.A., Inc., ___ S.E.2d ___, 2024 WL 5249823 (N.C. App. Dec. 31, 2024), reversing yet another order that a foreign defendant pay for translating its own documents:

[W]e conclude the trial court erred by requiring the . . . Defendants to create new documents in English of documents already provided that are in the Japanese language.  Rule 26 of our Rules of Civil Procedure allows a party to seek documents in the possession of the adverse party; it does not generally require the adverse party to pay for any said documents to be translated into the English language.  In other words, there is no duty to produce documents that do not exist.

Id. at *2 (citation omitted).  Sessoms was “persuaded by what we perceive to be the greater weight of authority in the United States that a party producing documents is not required to create new documents consisting of English translations of documents already provided.”  Id. at *3.  Sessoms cited eight cases exemplifying that “authority” – three of which were decided after our 2008 post.  That suggests that the post could use an update.

So here’s that update.

Continue Reading Still Nyet, Defendants Not Required to Pay for Translation of Documents
Photo of Eric Hudson

We write a lot about the learned intermediary rule. There are 50 state surveys and  summaries of helpful decisions, as well as numerous posts on state-specific decisions. We tracked the development of the rule in jurisdictions like West Virginia and Arizona, and we’ve generally been pleased to report positive developments. At the end of 2024, though, we flagged the Himes case from California as one of the ten worst decisions of the year based on its novel approach to warnings causation.  Given our criticisms of Himes, we found it both bizarre and troubling to see the case cited recently by an MDL court in Massachusetts applying Pennsylvania law. Our colleagues in the plaintiffs’ bar are clearly advocating to expand Himes into other jurisdictions, and the defense bar should be ready to counter those efforts.

Continue Reading Himes Makes a Sneak Appearance on the East Coast
Photo of Eric Alexander

This time of year, many of us are focused on the NFL playoffs.  For someone who watches the Super Bowl for the commercials or the halftime show, which team wins may not matter much.  For those devoted to a particular team, however, there is one possible result that will be truly satisfying.  The reality is that, the fans of 96.875% of the teams end the season with dissatisfaction and those of 92.857% of the playoff teams end the season with an excruciating loss.  Of course, there are those who can find something positive from a non-championship season.  Maybe the fan’s favorite team showed a big improvement over the prior season or the team finally found a franchise quarterback, either of which could lead to optimism for the next season, even though it will probably result in another non-championship.

In the world of product liability MDLs, way more than 99% of the cases filed end in dismissal.  In any given MDL, very few, or perhaps none, of the cases that the plaintiff lawyers file will result in judgment for the plaintiff.  Of course, the path to and nature of the dismissal will determine how the parties and their counsel feel about it.  A dismissal in connection with a settlement or a dismissal without prejudice will not be as dissatisfying for the plaintiff as a dismissal with prejudice after summary judgment or a defense verdict at trial.  For the defendants in an MDL, the reality is that satisfaction is affected by a range of economic considerations.  Managing an MDL where the vast majority of cases essentially sit around for years waiting for settlement requires a focus on the big picture as well as the case count.

Subject matter jurisdiction in a product liability MDL will almost always be based on diversity and the defendant will usually prefer that any given case within the MDL’s definition proceed in the MDL instead of in state court.  There is a clear gap in the proceeding statements.  A plaintiff with the same state of residence as the defendant(s) cannot stay in federal court based on diversity jurisdiction, and there is no special federal jurisdiction for MDL courts.  In In re Cook Med., Inc., IVC Filters Mkt’g, Sales Pracs. & Prods. Liab. Litig., MDL No. 2570, 2024 U.S. Dist. LEXIS 235780 (S.D. Ind. Nov. 14, 2024) (“Cook”), the defendants succeeded in getting the cases of eleven plaintiffs from Indiana, where the defendants are also based, dismissed without prejudice for lack of subject matter jurisdiction.  That is only part of the story.  The plaintiff lawyers admitted they did not have diversity jurisdiction for these cases when they were directly filed in the MDL and no other basis for keeping the cases in the MDL.  And they refused to dismiss the cases when requested.  So, the defendants sought Rule 11 sanctions in connection with the dismissals.  This was largely symbolic because they sought only the costs of bringing the motions to dismiss in nine cases and later reduced the request to $100 per case.  They apparently did not pursue other avenues for sanctions or costs, probably because Rule 11 allows sanctions to be awarded against an attorney or law firm, not just against a party.

The Cook court denied sanctions.  While the plaintiff lawyers had no argument about jurisdiction, their excuse for refusing to voluntarily dismiss when requested was that Indiana’s savings clause would not provide a year to re-file if the dismissals were voluntary.  The court thus saw the refusals to dismiss as not being sanctionable because they were “in the best interest of Plaintiffs, their clients.”  2024 U.S. Dist. LEXIS 235780, *4.  Here is the problem with that decision.  The cases were directly filed in the MDL in 2018 to 2020, several years after the MDL was established.  [We have tracked various rulings from and related to this MDL over the years, like here, here, here, and here.]  Subject matter jurisdiction is measured at the time the case is filed in or removed to federal court.  Plaintiff lawyers should know the law well enough to ensure that they do not file time-barred cases and/or cases in courts without jurisdiction.  There would have been no need to use the Indiana savings clause if the cases had been timely filed in Indiana state court originally or if the mistake of filing in the MDL had been discovered before the statute of limitations expired.  These cases were pending in the wrong court for four to six years before the savings clause became the excuse for forcing motions to dismiss to be filed.  When the court had to rule on those motions and the attendant obvious sanctions issue, the delay should not have been exculpatory.  After all, the Cook court did find that the plaintiff lawyers “should have realized that Plaintiffs’ cases did not fall within the diversity jurisdiction of this court,” a determination that would have applied at all points from before filing until the ultimate dismissal.  Id. 

Sanctionable conduct by a lawyer should not get a retroactive blessing if it persists long enough to hurt the lawyer’s clients.  Nor is the consideration that sanctions “can affect the reputation and creativity of counsel” usually going to be sufficient for an MDL to refrain from sending the message that filing frivolous cases and refusing to dismiss them is sanctionable.  Id. (quoting from Hartmarx Corp. v. Abboud, 326 F.3d 862, 867 (7th Cir. 2003)).  Indeed, because the savings clause would not resurrect a case that was untimely when filed, if any of these cases were filed in the wrong court after the statute of limitations had expired, then the much later refusal to dismiss served no interest of the lawyer’s client.

By contrast, just about any unsupportable claim or position advanced by a plaintiff lawyer could be said to be done in the “best interest” of the lawyer’s client, who presumably wants to win a big award or secure a big settlement.  That may be why Rule 11 has no provision that excuses unsupported representations to the Court—such as a jurisdictional allegation in a complaint—based on the interest of the party on whose behalf the representation is made.  It does specify, however, that any “sanction imposed under this rule must be limited to what suffices to deter repetition of the conduct or comparable conduct by others similarly situated.”  Fed. R. Civ. P. 11(c)(4).  In the context of an MDL, trying to deter the direct filing of cases without subject matter jurisdiction makes sense.

Photo of Stephen McConnell

Blair v. Abbvie Inc., 2025 WL. 57198 (W.D. Pa. Jan. 9, 2025), is, from the defense perspective, a favorable opinion dismissing (some with prejudice, some with leave to amend) all counts of the plaintiff’s complaint.   The opinion is a bit odd, in a semi, unintentionally-ironic sort of way, because it faults the plaintiff for not supplying enough information, while the court’s opinion does not tell us what the product is, except it is some kind of implant.  We’re guessing that the implant had something to do with the plaintiff’s eye, because the court observes in a footnote that the plaintiff included among the named defendants an eye care subsidiary … that did not actually exist.  Call it foreshadowing. In most of the complaint, there was no there there.

The Blair opinion follows Pennsylvania law to the effect that Restatement (Second) of Torts section 402A, Comment k requires dismissal of strict liability design claims.  Some W.D. Pa decisions bounced strict liability design defect claims on this ground and some did not. The Blair court sided with the better W.D. Pa. decisions, logic, truth, the American way, and an E.D. Pa. decision (Smith v. Howmedica Osteonics Corp., 251 F. Supp. 3d 844, 847-51 (E.D. Pa. 2017) – it is always delightful to see a sound decision from our home district) by applying comment k to bar the strict liability claims.  That same logic ended up also barring the implied warranty design defect claims, because the elements of the strict liability and warranty claims were “coextensive.” Dismissal was with prejudice because there is nothing the plaintiff could do to fix those claims. 

Sadly for the defendants, comment k did not bar the negligent design defect claim, nor any species of the manufacturing defect claims.  But sadly for the plaintiff, all the other claims in the complaint were inadequately pleaded. What the complaint said was purely formulaic. The allegations were “too broad and conclusory.” All the plaintiff tells us is that the product was “manufactured and sold by” the defendants, “that it was defective when it was sold, that it reached him without changes in its condition, and that he was injured after it was implanted.” Not good enough. The manufacturing claim did not state what went wrong or even how the product allegedly failed.  Such gaps/omissions meant that the claim did not even rise to the level of a dreaded “malfunction theory” claim.  

This inadequate pleading carried over to the plaintiff’s negligence-based claims, all of which were also dismissed. The plaintiff did not “identify the design defect, anything about the manufacturing process, or what information” the defendants should have given to the plaintiff’s medical providers. The Blair court helpfully gathers some good W.D. Pa. precedents, and they will be useful if you find yourself defending a case in the Pittsburgh area.  The judges will likely take the time to read such precedents, as they will not be distracted by the local professional football team, which has made its annual first round exit from the playoffs.

Meanwhile, the plaintiff will get the opportunity to try to do better.  So, next year, will the Steelers.  We predict that neither will succeed.

Photo of Michelle Yeary

Defendant in Beavan v. Allergan U.S.A., Inc., 2014 N.J. Super. Unpub. LEXIS 2898 (N.J. App. Nov. 21, 2024) made two solid arguments for summary judgment – preemption based on the FDCA’s recall regulations and plaintiff’s lack of admissible expert testimony.  The trial court rejected both.  The appellate court, however, saw the merit in the expert argument and that was all defendant needed to secure a dismissal. 

Defendant manufactures a prescription drug that is used in the treatment of various eye diseases.  The drug is administered by injection.  A routine product inspection revealed that approximately 2.2% of the units of a particular lot of the drug contained a silicone particulate which could be dispensed when the drug was injected.  Defendant sought FDA approval to send out a Dear Health Care Provider letter (“DHCP letter”), but the FDA concluded the issue was not a “safety concern.”  Instead, and with the FDA’s approval, defendant issued a drug recall and a notice to physicians that there was a low risk of a mild reaction.  Id. at *2-4.  Plaintiff received an injection of defendant’s drug from the lot that was recalled.  After which she was diagnosed with retinal detachment.   Plaintiff filed her complaint alleging claims for design, manufacturing, and warning defects under New Jersey’s Products Liability Act. 

Defendant’s preemption argument centered on the FDA’s recall regulations which it contended preempted the state from regulating prescription drug recalls.  The court disagreed finding that the voluntary nature of the recall defeated defendant’s implied preemption argument.  Because a manufacturer can add a warning to its labeling, which includes via DHCP letters, without FDA approval, defendant was not foreclosed from warning physicians about the manufacturing defect and therefore, there was no preemption under Wyeth v. Levine, 555 U.S. 555 (2009), and its progeny.   Further, while the NJ PLA affords defendants a presumption that their FDA-approved labeling is adequate, the presumption is rebuttable with allegations that the manufactured acquired knowledge of harmful effects after the sale of the product and failed to disclose them.  In other words, NJ recognizes a post-sale duty to warn.  Here, defendant’s DHCP letter was sufficient to rebut the adequacy presumption and therefore a sufficient basis on which to assert a non-preempted failure to warn claim.

But that is where the bad news ends because plaintiff’s experts took a leap of faith and assumed that the injection plaintiff received contained the defect.  You know what they say about people who assume.  While both of plaintiff’s experts testified that the silicone particulate caused plaintiff’s injuries, neither offered any evidence that the “injection plaintiff received was defective and no evidence of a particulate in her eye.”  Id. at *34.  Therefore, plaintiff’s theory of causation was “based on evidence that does not exist and would leave a jury to speculate whether there was ever a particulate in the applicator or particulate injected into plaintiff’s eye.”  Id.   New Jersey’s net opinion rule prohibits the admission of an expert’s conclusions that are not supported by “factual evidence or other data.”  Id. at *32. 

Not only did Plaintiff’s experts based their conclusions on an unsupported assumption, they failed to rule out that plaintiff’s injuries could have been caused by various other factors—such as a different silicone insert used in her treatment that became dislocated at the same time she suffered her injury or that retinal detachment is a risk of some of plaintiff’s pre-existing and underlying eye diseases.  Id. at *34-35.  The court also acknowledged that New Jersey has recently adopted the Daubert factors for expert admissibility and that plaintiff’s experts would not pass under that standard either.  Id. at *35-36. 

The case was reversed and remanded for summary judgment to be entered on the grounds of no expert evidence of specific or general causation.  We would have liked a better preemption outcome, but we are not greedy.  A lack of causation is good enough for us any day.

Photo of Bexis

This blog has long encouraged defendants in prescription medical product liability litigation to seek relevant ediscovery from plaintiffs.  We even have an ediscovery cheat sheet with almost 250 favorable decisions either allowing defense-side ediscovery in personal injury cases or else sanctioning plaintiffs for spoliating sought-after electronic data.  But we confess, we’ve been focused so firmly on social media and smartphones, where ediscovery from plaintiffs originated, that we have ignored the rising popularity of fitness trackers, Fitbits, smart watches, smart rings and similar devices (even clothing) being marketed to people who may eventually become plaintiffs.  These products create a great deal of health-related (and other) information that is of obvious relevance in mass (and other) tort litigation.

What we found is that surprisingly few defendants seem to be seeking this type of information – at least there are very few decisions involving discovery of these devices.

Continue Reading Ediscovery for Defendants – The New Frontier