Taking a sad song and making it better, St. Jude Medical recently scored double-barreled preemption victories in implantable cardioverter defibrillator (“ICD”) lead litigation. On consecutive days, federal courts in Minnesota, Pinsonneault v. St. Jude Medical, Inc., 2014 WL 2879754 (D. Minn. June 24, 2014) (Riata), and New York, Franzese v. St. Jude Medical, Inc., 2014 WL 2863087 (E.D.N.Y. June 23, 2014) (Durata) issued favorable decisions. In the post-Daimler AG v. Bauman, 134 S.Ct. 746 (2014), personal jurisdiction landscape, the Pinsonneault decision is particularly sweet, because St. Jude Medical is based in Minnesota.
The Pinsonneault Riata plaintiffs alleged that their leads were subject to a recall. 2014 WL 2879754, at *1. However, these plaintiffs wisely did not dispute that PMA preemption applies to recalled products.
As is usual in lead litigation (see, e.g., Kemp v. Medtronic, Inc., 231 F.3d 216 (6th Cir. 2000), one of Bexis’ cases), plaintiffs in both cases tried to fit their allegations into the rubric of manufacturing defect. That’s not a bad strategy for the other side to follow, as some manufacturing defect allegations can at first blush be made to look like a plausible “parallel violation” claim – simultaneously being both a traditional state law claim and an FDA violation (of some applicable, and usually vague, FDA Good Manufacturing Practice (“GMP”)). Indeed, the Pinsonneault plaintiffs had no choice, as all their non-GMP-related claims had already been dismissed as preempted by an earlier decision. See Pinsonneault v. St. Jude Medical, Inc., 953 F. Supp. 2d 1006 (D. Minn. 2013).
Preemption nonetheless applied in both cases, with the key being that the defendants were able to demonstrate that the plaintiff’s claimed GMP “requirements” in fact were fanciful.
In Pinsonneault, the company unfortunately – and we would say unfairly – had to bear the expense of discovery to pin down the plaintiffs and their experts on their GMP theories on summary judgment. Even so, there was considerable back and forth as the plaintiffs repeatedly tried to expand discovery beyond the court-imposed limitation to preemption-related issues. 2014 WL 2879754, at *5-7. In Franzese, the defense managed to prevail on motion to dismiss. 2014 WL 2863087, at *1-2.
As is usual in GMP-related manufacturing defect cases, the Pinsonneault plaintiffs eventually had to come to grips with the vagueness of the FDA’s language. The particular GMP, 21 C.F.R. §820.70(a), simply “requires device manufacturers to develop and document manufacturing processes ensuring that devices conform to their specifications.” 2014 WL 2879754, at *7. Thus, the Pinsonneault
plaintiffs tried a bootstrap approach that looked to processes the defendants had put in place – and claimed transubstantiation − that these internal processes somehow became FDA requirements. They
lost. Boy, oh boy, did they lose:
requirements for purposes of §360k(a). The absence of such case law is not surprising, as such a holding would plainly be at odds with the fact that the [GMP] purposely does not prescribe
specific manufacturing processes in order to give manufacturers a measure of flexibility. . . .
regulated under that provision − not just Class III medical devices subject to the PMA process − would be subject to detailed and specific federal requirements that would preempt additional or different state requirements.
2014 WL 2879754, at *7-8 (citations omitted) (emphasis added).
After some squirming and gyrating, the plaintiffs in Pinsonneault eventually were confined to five manufacturing defect theories: (1) lack of uniform insulation thickness (shades of Kemp, 231 F.3d at 230-32 (addressing similar manufacturing defect claims in lead litigation)); (2) lack of controlled, uniform crimping force; (3) not using approved methods and specifications for curing the insulation; (4) lack of consistent “lubricious interface” between the inner and outer insulation; and (5) improper sterilization. 2014 WL 2879754, at *4.
Identifying the specifics of a plaintiff’s moving target defect claims is half the battle – and in Pinsonneault and the discovery fights needed to do that lasted over a year.
Here’s what happened to these purported “defects”:
Thickness – “The only limit is the outer-body diameter requirement for the entire lead, but that simply sets a maximum diameter for the insulation; it does not require that the insulation be uniform or that it be of any particular thickness.” Id. at *9. “Nor is it true, as plaintiffs argue, that the prohibition on nicks or other defects greater than 20 percent of the wall thickness necessarily means that the insulation must be a particular thickness. A percentage, by its nature, can apply to any number.” Id. The FDA could request samples, but didn’t do that here. “[P]laintiffs cite no evidence that St. Jude provided a sample lead with its PMA application, much less evidence that any such sample had a uniform insulation thickness.” Id. at *10.
Crimping – “[T]he required crimp depths can be achieved with a range of forces and that there is no requirement for a controlled, uniform degree of force when crimping. . . . Because there is no evidence of any requirement for a controlled, uniform degree of force when crimping, plaintiffs’ claims that the leads were defective because of the failure to use a controlled, uniform degree of force are preempted.” Id.
Curing – “[P]laintiffs admitted at oral argument that they have no evidence of any federal requirements as to curing.” Id.
Lubricious interface − “[P]laintiffs admitted at oral argument that they have no evidence of any federal requirements as to . . . lubricious interface.” Id. Not only that, but the issue simply didn’t exist. “[T]he only evidence in the record is that the leads do not have a lubricious interface between the inner and outer insulation.” Id.
Indeed, the only claim that survived preemption in Pinsonneault was the last one, about sterilization, and it survived only because the defendant’s response – yes, we followed proper sterilization procedures – wasn’t a preemption based argument, and thus fell outside the limited scope of the current proceeding. 2014 WL 2879754, at *4. The dismissal was without prejudice to the company re-raising its sterilization arguments at a procedural-proper time. Id.
But even as to the sterilization theory, the plaintiffs received a judicial warning. “[U]nless” plaintiffs have “evidence that what St. Jude says is untrue,” they “should voluntarily withdraw their sterilization claims and not require St. Jude to bring (and this Court to rule on) another summary-judgment motion.” Id. at *4 n.3 (citing Fed.R.Civ.P. 11(b)(3) (prohibiting the advocacy of a factual contention that lacks “evidentiary support”)). To which we would add, see discussion of curing and lubricious interface.
Pinsonneault took as long as it did because the court tolerated vague GMP manufacturing defect claims at the pleading stage. As previously mentioned, some courts “let it in” but others “let it out.” Franzese demonstrates what happens to the same sort of claims in a jurisdiction (the Eastern District of New York) that does not allow plaintiffs get away with vague allegations about supposed violations of federal manufacturing requirements subject to multiple possible interpretations. Plaintiff’s vague allegations in Franzese about “unsanctioned adulterations . . . in violation of CGMPs” (the “C” is short for “current”) simply didn’t cut it.
Franzese, 2014 WL 2863087, at *4 (citations omitted).
In Franzese, plaintiffs cited an FDA warning letter in their ultimately futile effort to invent some federal violation. The warning letter, however did not even apply to the device at issue (it was issued three years later, id.), and it was also vague – just like FDA GMPs are vague. It “simply stated that the Durata lead was considered adulterated . . . because ‘the methods used in, or the facilities or controls used for, their manufacture, packing, storage, or installation’ are not in conformity with CGMPs. The specific CGMPs identified, though, do not have any direct implications on how or why the Durata lead prematurely deteriorated.” Id. at *5 (citation omitted). The warning letter could rise no higher than the vague GMPs it cited:
Id. (citations omitted) (emphasis added).
Preemption of the remaining claims in Franzese was easy. The warning letter addressed manufacturing, not design issues. Without any alleged federal violation, the design defect claim also fell to preemption. 2014 WL 2863087, at *6. Warning claims likewise failed. Plaintiffs again pointed to the warning letter, but the extent the letter had anything to do with warnings, it was irrelevant because it was dated three years after the relevant events. Id. Warranty claims failed for failure to state the purported warranty (express) and because the warning letter could not be equated with a “finding that the devices were unsafe or unfit for their ordinary purpose” (implied). Id. at *7. Negligence claims failed for the same reason as strict liability claims. Id. at *8.
Pinsonneault and Franzese demonstrate how PMA preemption goes a long way to making ICD lead claims against St. Jude Medical lost causes, which we suppose is hagiographically appropriate. Sha na-na-na; sha na-na-na, hey-hey-ay, good bye.” So what if it’s not Hey Jude (it’s Steam) – it’s also appropriate.