This post is from the non-RS side of the Blog.

Consider the following scenario.  A fifty-two-year-old woman has end-stage left ventricular heart failure despite medical care and the latest medications.  Her prospects for a heart transplant or the implantation of a left-ventricular assist device to prolong her life are limited, including by financial considerations.  She chooses to enroll in a clinical trial for people like her, is accepted after a rigorous informed consent process, and is randomly allocated to the investigational device group in the study.  The investigational device is a cutting-edge magnetically levitated centrifugal-flow left ventricular assist device, whereas the other option in the study is a mechanical-bearing axial-flow left ventricular assist device.  Her investigational device is implanted at a premier medical institution.  All the costs of her care in connection with the study are borne by the sponsor of the study, a medical device company that also makes some of the axial-flow devices on the market.  A few years before her enrollment, FDA approved an Investigational Device Exemption for her clinical trial.  Even though the new device is intended for short-term use in a heart failure patient or as a short-term bridge to a heart transplant, the woman lives another three years with the magnetically levitated centrifugal-flow left ventricular assist device keeping her heart functioning.  Along the way, FDA approves the device’s Premarket Market Approval application, including the device’s design, warnings, and other labeling.  Yet the study continues to provide the woman with free medical care related to her implanted device.  Also, the results of the clinical study are published three times in the New England Journal of Medicine, with the third publication noting that “a fully magnetically levitated centrifugal-flow left ventricular assist device was associated with less frequent need for pump replacement than an axial-flow device and was superior with respect to survival free of disabling stroke or reoperation to replace or remove a malfunctioning device.”  In laymen’s terms, the investigational device was a great success.  About a month after that third article, the woman was alerted to a malfunction in her implanted device, so she called the number she had from her investigators/providers at this premier medical institution.  Unfortunately, the woman died that day, which her daughters blamed on the level of training of the provider with whom the woman spoke on the day she died.  Instead of being thankful for the three extra years with their mother and the free care she had received, the daughters sued.  In addition to the hospital and investigator, they sued the manufacturer/sponsor in California state court under a number of product liability theories.

This is what happened in Galdamez v. Abbott Cardiovascular Sys., Inc., No. CGC-21-591082 (Cal. Super Ct. May 7, 2026), slip op., with a little color added from a few minutes of fishing around on reputable sites on the internet.  It should be obvious to Blog readers and anyone knowledgeable about preemption, that most of the plaintiffs’ claims would be preempted.  Design and warnings claims against devices with a PMA or IDE approval at a relevant time are preempted.  Parallel claims are an overblown fallacy.  Plaintiffs will almost never have the device- or lot-specific evidence to sustain a manufacturing defect claim, generally making design arguments and labeling them manufacturing arguments.  We saw all of that in Galdamez.  We also saw that the plaintiff failed to dispute any of the manufacturer’s facts according to the California summary judgment procedure and failed to respond to some of the manufacturer’s arguments at all.  By the time the court ruled on the manufacturer’s motion for partial summary judgment, about five years had passed since the case was filed.  Regardless of why it took so long or why defendants held off on seeking complete summary judgment—the court even said the express preemption claim that was not raised in the motion would be preempted once challenged—the remnants of the case limp along when the case never should have been brought against the manufacturer/sponsor in the first place.  (The decision does not provide enough information to weigh in on whether plaintiffs could show substandard care by the decedent’s healthcare providers, but that would not create liability for the manufacturer/sponsor anyway.)

So why are we discussing a state court partial summary judgment ruling?  We certainly do pay attention to cases addressing the liability of clinical trial sponsors.  That has been an interest since the Blog’s infancy.  We also like to see what wacky arguments plaintiffs make as they try to dance between express and implied preemption.  We also think this is a good example of what is wrong with the “duty to innovate” claim presently before the California Supreme Court.  Taking the last first, the plaintiffs did not assert that the manufacturer/sponsor has breached a supposed “duty to innovate” and the court did not address it.  After all, this device was the first of its kind.  Pretend, though, that a different plaintiff was suing over one of this defendant’s marketed axial-flow devices that was implanted at the same time as the implant in the Galdamez plaintiffs’ decedent.  It could even have been implanted within the Galdamez plaintiffs’ decedent’s clinical trial as part of the established therapy arm of the trial; this was a non-inferiority trial with, of course, no ethical possibility for a placebo comparator.  At what point in time—interim results, PMA submission for the new device, PMA approval, or two years after PMA approval, to name four options—would the ludicrous “duty” require the manufacturer to pull its axial flow devices in favor of the centrifugal flow device?  Given the position of the Bartlett court (and, apparently, the Galdamez court) that any stop selling theory for an FDA-approved medical product would be preempted, why would any court entertain creating a novel but invariably preempted claim?  As we run through what plaintiff threw at the innovative device in Galdamez, we suggest keeping in mind how other plaintiffs would use the innovative device in a suit over an older device that was still marketed and widely used.

First up were plaintiffs’ design defect claims.  Plaintiffs contended that the design approved as part of the IDE and later the PMA was defective because of its purported propensity to blow a fuse.  Because of the timing, the preemptive effect would have come from the IDE approval, so plaintiffs contended that IDEs should be treated like 501(k)s under Lohr and have no preemptive effect.  The court disagreed based on a California case, Robinson v. Endovascular Techs., Inc., 190 Cal.App.4th 1490 (Cal. Ct. App. 2010), and two federal appellate decisions, Slater v. Optical Radiation Corp., 961 F.2d 1330 (7th Cir. 1992), and Martin v. Telectronics Pacing Sys., Inc., 105 F.3d 1090 (6th Cir. 1997).  We could quibble with the court saying it was bound by California authority on a Supremacy Clause issue and its omission of plenty of additional authority supporting IDE preemption of design claims, but the best plaintiffs could do was complain that the California case is old and an outlier—which it is not.

Next up was a claim for failure to recall, which is pretty silly in the context of a pre-market clinical trial.  During briefing, Plaintiffs tried to shift to a “failure to report to FDA” theory based on the horrible Stengel decision (that turns on a non-existent state law claim).  The original claim was both abandoned and also preempted under Buckman, both because FDA has exclusive enforcement over withdrawing an IDE or recalling a marketed device.  During argument, Plaintiffs tried to shift to a failure to warn the clinical trial investigators and participants theory not asserted in their live complaint, but that also would have been preempted.

Plaintiffs also offered implied warranty theories, which failed under state law for lack of privity and were preempted for the same reasons that the design defect claims were.  We pause to ponder the assertion that a free, life-prolonging device received in connection with a randomized clinical trial where the patient knew in advance that she had a 50% chance of getting an investigational device could ever breach an implied warranty of merchantability.

Plaintiffs also tried to make the sponsor liable for the alleged wrong emergency number provided to plaintiffs’ decedent by the medical institution where she got her three years of free care in connection with the clinical trial.  Lacking any “authority for the proposition that a device manufacturer . . . owes a duty to a patient in a clinical trial to provide her with an emergency number,” plaintiffs were left to argue that the clinical trial was a joint venture under California law.  The clinical trial documents said that it was not and there were no other features of a joint venture, such as shared profits.

While it is possible that a clinical trial subject could receive substandard medical care in connection with her voluntary participation in a clinical trial that might lead to liability for treaters, when the trial involves an IDE device, the sponsor is almost never going to have liability.  After five years on the docket, there is no indication that Galdamez is the rare exception, and it is time for the sponsor of a trial that helped bring an innovative device to market to be out of the case.