Photo of Eric Hudson

We’ve been posting about decisions from In re Smith & Nephew Birmingham Hip Resurfacing (BHR) Hip Implant Products Liability Litigation, MDL 2775, since 2018.  Its preemption ruling on defendant’s motion to dismiss made the list of ten worst decisions from 2018, and a subsequent preemption decision reflected more MDL madness.   Things improved when the court began addressing causation at summary judgment, and pretty soon cases were falling like dominoes.  Today’s decision from the MDL, Williams v. Smith & Nephew, Inc., 2024 WL 99542 (D. Md. Jan. 8, 2024), continues that positive trend.

The patient in Williams underwent a hip resurfacing with the defendant’s device and complained of complications about six years later.  Metal testing showed elevated levels of cobalt in the patient’s body, leading to replacement of the device with another product.  The patient then died of apparently unrelated causes. Two of his family members pursued his claim.  The court previously dismissed most of the plaintiffs’ claims in omnibus rulings, so the only claim left here was negligent misrepresentation.  

Plaintiffs rooted their misrepresentation claim in allegations about a “learning curve” for physicians to implant the device with lower revision rates. One of the inventors of the device stated that his first 1,000 implants with the device had a ten year failure rate of 4.7%, but his next 3,000 implants had a failure rate that was 5% better. Plaintiffs argued that this “learning curve” of 1,000 cases should have been communicated to the implanting physician.

Presumably in response to typical plaintiff-side questions, the implanting physician testified that knowing about the learning curve would have been “meaningful,” and that he would have wanted to know “as much information about the risk of failure and revision surgery as possible.” Id. at *2 (we’ve previously posted that such testimony is insufficient to establish causation). The implanter had attended training with the manufacturer to learn the implant procedure.  During that training he was informed that the failure rate of the device would be between one and three percent at ten years after implantation.  Plaintiffs claimed this one to three percent failure rate was a misrepresentation since it conflicted with the learning curve statements by of one of the inventors.

The Court had none of it:

Even assuming that the failure to describe the learning curve is not a preempted attack on FDA-approved information . . ., the [Plaintiffs’] argument relies on impermissible speculation and an untenable chain of inferences, and therefore fails to make out a viable negligent misrepresentation claim.

Id. at *5. The Court premised its ruling on several grounds.

First, there was nothing to show that anything related to the learning curve was a proximate cause of the alleged injuries.  Plaintiffs contended that the injuries were the result of increased levels of cobalt released during the device’s normal function—an alleged inherent defect in the product. Their claim was not premised on surgical error, so it had nothing to do with any learning curve for the implanting physician.  There was simply no evidence to link the elevated levels of cobalt and plaintiffs’ claims to any surgical error or the learning curve evidence.  Moreover, there was evidence in the record that the implanter properly implanted the device.

Second, plaintiffs did not show that the device manufacturer was aware of the inventor’s views about the learning curve at the time it trained the implanter and made the representation regarding a one to three percent revision rate. The statements of the inventor were introduced into the record by a video of an interview played at another witness’s deposition. But the interview took place several years after the implanter’s training. The court pressed plaintiffs’ counsel on this point during oral argument, and the best counsel could do was argue that the manufacturer had constructive knowledge of the views of the inventor—essentially arguing that, because the manufacturer knew the procedure was new, it must also have known about the learning curve. The court’s response was that “[i]t is not reasonable to infer knowledge of specific data from general concerns.”  Id. at *6.

Third, the Court rejected the plaintiffs’ argument that the deceased would have foregone implantation of the device if the learning curve had been explained to him.  Plaintiffs pointed to evidence that the deceased was generally suspicious of medical procedures and he conducted his own research about hip implants prior to agreeing to the surgery.  The manufacturer countered that the deceased trusted his implanter and that there was no evidence the implanter would have changed his recommendation to the patient based on the learning curve.  Noting a lack of evidence regarding the implanter making any change to his recommendation and the speculation inherent in plaintiffs’ position, the Court rejected this argument.

Finally, the Court rejected the plaintiffs’ argument that the manufacturer should have disclosed learning curve data to the implanter once it became aware of the information.  Plaintiffs contended that, if the manufacturer had disclosed this information to the implanter, the implanter  would have conveyed that information to the patient, and the patient would have then sought treatment for device failure sooner than he did. The Court viewed this chain of events as too speculative to support a negligent misrepresentation claim. If anything, this would be a failure to warn claim—which the Court previously dismissed.  Chalk up another “W” in this MDL.