Two weeks ago, we decried the pattern that some courts follow in allowing shifting slates of boilerplate allegations to cases to discovery. The decision in Corrigan v. Covidien LP, No. 22-cv-10220, 2022 U.S. Dist. LEXIS 210296 (D. Mass. Nov. 21, 2022), reminded us of another of our post-TwIqbal pet peeves: when courts treat sweeping legal conclusions as if they were plausible factual assertions. The basic allegations in Corrigan were that the plaintiff’s surgeon used defendant’s surgical stapler to perform an anastomosis—reattachment of two parts of the digestive tract—in connection with removing part of his sigmoid colon (for unspecified reasons, but often diverticulitis or cancer) and the anastomosis later leaked, leading to further surgical intervention. As we said two weeks ago, medical device manufacturers are not insurers. That makes sense because surgery on humans, even done by the best surgeons, in the best hospitals, and with the best devices and equipment, has less than a 100% success rate. Anastomoses leak, infections develop, hernias recur, patients report post-operative pain, and all manner of complications and less than optimal outcomes occur. A common refrain when scientists are presented with a surgical study reporting no complications or failures is that the study was too small, too short, and/or insufficiently rigorous. Thus, a common procedure with a very high success rate will still generate large numbers of reported failures, like anastomoses that leak. This is part of why rates, and particularly comparative rates, provide more useful information about devices and surgeries than do gross numbers.
Even more authoritative sources than this Blog agree with us. FDA, for instance, makes clear that data from its Manufacturer and User Facility Device Experience (“MAUDE”) database “is not intended to be used either to evaluate rates of adverse events or to compare adverse event occurrence rates across devices.” The reasons for this are not a secret: “The incidence, prevalence, or cause of an event cannot be determined from this reporting system alone due to under-reporting of events, inaccuracies in reports, lack of verification that the device caused the reported event, and lack of information about frequency of device use.” Directly stated, “[t]he submission of an MDR itself is not evidence that the device caused or contributed to the adverse outcome or event.” FDA’s description of the MAUDE database also makes clear in a number of places that the data may be outdated or incomplete for various benign reasons.
Back to Corrigan. Keeping in mind that an anastomosis after a sigmoidectomy is supposed to allow for the passage of fecal matter towards the rectum without leakage, the core allegation was lack of efficacy, which is rarely a fit for product liability claims. Yet, plaintiff offered up the typical claims of design defect, manufacturing defect, and warnings defect—each of which Massachusetts labels “breach of warranty” but treats like strict liability—negligent warnings, negligent design, negligent marketing, negligent misrepresentation, and deceptive trade practices under a Massachusetts statute. We will cut to the chase and say the Corrigan court correctly dismissed the unsupported design and manufacturing claims but looks like it fell prey to some flimflam on the rest, which were really just the same warnings claim in slightly different outfits.
The good part first. Plaintiff alleged that the post-operative anastomotic leak “indicated that the surgical stapler used during the procedure failed to completely seal the tissue.” Id. at *3. We do not have the complaint or briefs to review, but that does not look like a factual allegation on which a claim could be built in the era of TwIqbal. To us, that is boilerplate, if not a legal conclusion, equating to “the surgery failed to achieve its desired result, so it must have been because the device failed to perform.” Surgery is a poor candidate for res ipsa loquitur. How the device failed to perform, why it failed to perform, and why a failure to perform caused a compensable injury also need to be alleged. The court came at it from a different angle, but these omissions were why the manufacturing and design claims were dismissed. On the former, plaintiff did not “allege what that [incorrect manufacturing] process was or how the device ‘deviated’ from the intended design in a way that rendered it unreasonably dangerous and unfit for its ordinary purpose.” Id. at *9. Pointing to the recall of different surgical staplers made by defendant did not help because plaintiff did not identify “what the particular defect was in this device at issue where the issues in the other staplers were different from each other and allegedly caused by different issues.” Id. at *11. Similarly, plaintiff did not “point to a specific design defect in the EEA31 stapler” and did not come close to pleading a safer alternative design, offering the most conclusory of allegations that the “foreseeable risks of harm posed by the products at issue could have been reduced or avoided by the adoption of a reasonable alternative design.” Id. at *13 & 15. Massachusetts law required plaintiff to identify “‘some unique feature of design’ of the EEA31 stapler that makes it ‘more dangerous than other[]’ staplers.” Id. at *14. The counts for negligent design and negligent marketing were both actually purported negligent design claims and required a design defect, which plaintiff did not articulate. Id. at *18-19.
On the warnings claims, some of the boilerplate language quoted above is instructive. When plaintiff wrote “foreseeable risks of harm posed by the products at issue,” it seems to us that was because plaintiff had no specific risk to complain about—beyond efficacy below 100%—and his issues were with surgical staplers as a whole. The generic liability allegations—and there were no case-specific liability allegations as far as we can tell—bear this out. Everything is about unspecified “surgical stapler malfunctions” and how they were reported to FDA, nothing specific to efficacy, risks, design, or warnings for this particular stapler. Id. at *4-5. The closest allegations were those about the defendant’s general FDA reporting practices and its recall of two other staplers. Plaintiff’s issue with reporting, though, was more an attack on FDA that was unsupported by factual allegations. We will try to boil it down. Plaintiff claimed 1) the defendant used the Alternative Summary Reporting Program (“ASR”) “instead of publically reporting each incident to” MAUDE, 2) that surgeons relied on MAUDE to learn about the comparative risks of surgical staplers, 3) surgeons were deprived of unspecified “important” information, 4) FDA ultimately scrapped the ASR system because of “misuse,” 5) FDA sent a letter to healthcare providers on surgical staplers, and 6) FDA reclassified them from class I to class II. (The irony of a plaintiff emphasizing upregulation to class II is not lost on those of us who have defended litigation over transvaginal mesh and other class II devices.)
Several things leap out at us about this theory for failure to warn, misrepresentation, and deceptive trade practices. First, it included nothing about the actual labeling for the device. It is hard to imagine how a claim that the manufacturer of a prescription medical device, particularly one used during surgery, could be pleaded without saying something about the written materials accompanying the device. For instance, if the labeling accurately described the use of the device, its benefits, and its risks, making it clear that there was a non-zero chance of leaks after an anastomosis with this particular stapler, then that adequate labeling should be the end of the story. Second, regardless of the format, adverse events get reported to FDA, not to potential prescribing surgeons. A warnings claim based upon what is said or not said to a third party should require something rather unusual to connect to the plaintiff’s case. For more on failure to report claims see our 50-state survey, here.
Third, the contention that all surgeons, and necessarily plaintiff’s surgeon, rely on gross number of all reports for all surgical staplers or anything else in MAUDE to decide how to perform an anastomosis is not a specific or plausible factual allegation. As we noted above, FDA makes clear that MAUDE does not allow for reliable conclusions on comparative risk, causation, or even that a search will yield accurate gross numbers. Relatedly, the hypothesized information gap from the use of the ASR option, to the extent it is a factual allegation at all as opposed to a boilerplate conclusion, is also implausible. According to FDA, which should know, the ASR system was not discontinued because of misuse by manufacturers. Reporting adverse events through ASR required an FDA decision on an exemption application:
Under this program, manufacturers of certain devices could request an exemption from the requirement to file individual medical device reports for certain events that were well-known and well-established risks associated with a particular device and to instead submit quarterly summary reports of such events. Since the program’s inception in 1997, the FDA granted 108 such exemptions to individual manufacturers for certain well-known events associated with specific devices, which were often already described in the product labeling available to health care professionals and patients. The ASR Program allowed the FDA to more efficiently review reports of well-known, well-understood adverse events, so we could focus on identifying and taking action on new safety signals and less understood risks.
Plaintiff may not have liked the ASR option while it existed or FDA’s decisions on exemptions, but the whole idea of a treating “well-known and well-established risks” differently runs contrary to plaintiff’s warning theory.
Fourth, and stop us if you have heard this one before, a federal court sitting in diversity cannot just create a novel state law cause of action. We saw no indication that Massachusetts recognizes a claim for failure to warn FDA. The Corrigan court certainly did not make an Erie prediction of whether Massachusetts would adopt such a cause of action. (Our 50-state survey strongly indicates that it would not.) Even in the context of the negligence claim, which requires a duty, the court failed to understand that the recognition in Massachusetts law that the duty warn about a prescription medical device runs to physicians meant there was no recognized Massachusetts duty to warn FDA.
Fifth, because the duty to report adverse events to FDA arises solely under federal law, Buckman prevents plaintiff from recovering based on a failure to report adverse events, whether by use of the ASR or otherwise. Moreover, because FDA controls the MAUDE database, any “warnings” relayed to surgeons through MAUDE requires actions by FDA. Under Mensing and Bartlett, that means plaintiff’s claim predicated on how the manufacturer reported adverse events that might have been seen by physicians who happened to search the MAUDE database should be preempted. Preemption was not raised in Corrigan, though.
Last, as we alluded to above in connection with an alleged claim based on a failure to warn a third party, there is nothing approaching proximate cause for an alleged failure to warn based on how adverse events were reported. Based on the opinion, the complaint does not even identify plaintiff’s physician by name, let alone include allegations about how her or his decision to use this particular surgical stapler in connection with plaintiff’s anastomosis might have changed based on some difference in the reporting of unspecified adverse events to FDA. Plaintiff had a portion of his sigmoid colon removed, so there was going to be an anastomosis and surgical staplers are the way the vast majority of those were being done in the US in 2019 (and today). There was no factual allegation suggesting a change in the choice of a surgical stapler in general or this one in particular based on how many adverse events were reported, or how they were reported, for surgical staplers.
The Corrigan court’s analysis pretty much missed all of this by blindly accepting plaintiff’s legal assertions and boilerplate as if they were plausible factual allegations. With our long lead-in, we will try to be brief in in describing the court’s actual analysis. “Plaintiffs plausibly allege that Defendants failed to provide Corrigan with ‘proper warnings on instructions’ so that he could avoid the danger associated with the EEA31 stapler.” 2022 U.S. Dist. LEXIS 210296, *15. Setting aside that the duty to warn did not run to plaintiff, the warnings provided were not discussed at all, and the dangers of the device were never described as higher than or different than the well-known risks of using any surgical stapler for an anastomosis, the plaintiff’s allegations were obviously insufficient even as the court described them. “Plaintiffs allege that Defendants reported adverse events through the ASR Program that . . . led to a knowledge gap where surgeons were unaware that the surgical stapler malfunction rate was so high.” Id. at *17. Particularly as this is about all surgical staplers and the “malfunction” in this case was never specified, this allegation should have been a non-starter. Similarly, the court accepted the conclusory statement that “Corrigan’s surgeons would not have had access to EEA31 staplers’ malfunction rate” as a substitute for factual assertions supporting proximate cause for failure to warn. At best, that might be part of proximate cause, but certainly not enough by itself.
The acceptance of these same statements and allegations as pleading negligently misrepresentation was probably worse. After concluding that the heightened pleading under Rule 9(b) applied to this claim, in the next breath, the court accepted general allegations about FDA reporting somehow affecting what was in MAUDE such that surgeons in general were not “fully informed of the risks that the devices presented.” Id. at *20-21. This does not begin to allege fraud on plaintiff’s surgeon in connection with this particular device and does not even touch on proximate cause. The deceptive trade practices claim survived simply because the failure to warn claims did and the FDA reporting decisions were allegedly made in Massachusetts. Id. at *24-25. Perhaps with some more education on how FDA reporting really works, the defendant will have a shot to get rid of this case based on preemption. It would be a shame if the occurrence of a “well-known and well-established” risk of surgery, without anything suggesting an actual increased risk with the device used in the case, equated to liability. After all, as the court recognized, plaintiff could not even articulate some particular issue with the device’s design or manufacture.