The article indicates that 3D printing of medical devices is already a reality. 3D printing is currently being used with implants where sizing is both critical and extremely patient specific – the article mentions tracheal splints, skull plates, hip cups, spinal cages, knee trays, and dental reconstruction devices. Here’s a cool slide show that we found on the FDA’s website showing some of them and discussing relevant 3D printing techniques. Also according to the article, the FDA doesn’t have any specific 3D printing guidance available – which we confirmed on Westlaw (no mention of “3D printing” or “additive manufacturing” in any FDA guidance).
We did find some interesting discussion on the FDA’s website when we searched for “3d Printing.” We read this description of 3D printing in the medical device context at a recent FDA public workshop:
So in general the design process includes the patient image acquisition, the image quality control and segmentation, patient modeling and anatomical definitions, preoperative planning and approval, guide design and patient match, feature definitions, and finally the guide construction. Each of these steps has the potential to employ proprietary and off-the-shelf software, and the software may be used by the manufacturer or even the end user. And software documentation is necessary, as the design of the patient matched guides is not static, it is essentially redesigned for every patient, and so we need to document the software that is being used in each stage.
October 8, 2014 Workship Tr. at 10. The critical interaction is between patient-specific imaging and the software on the 3D printer that incorporates the dimensions provided by that imaging into the software that describes whatever medical device is being produced. The primary regulatory suggestion seems to be to use the FDA’s existing guidances regarding software in dealing with 3D printing.
But we’re more about product liability than FDA regulatory compliance. What are the product liability implications of 3D printed medical devices? We’re limiting our discussion to medical devices, because 3D printed drugs would require molecular manipulation that we doubt is available yet.
The most basic products issue is, “who is the manufacturer/supplier?” As we discussed at length, here, the overwhelming national rule to date has been that hospitals are not strictly liable as “suppliers” of prescription medical products. According to the article, though, the 3D printers that produce medical devices are usually owned by hospitals. Does that potentially make them manufacturers? We’re sure plaintiffs will make that argument, but Restatement §402A requires a bit more than that, specifically that the defendant seller be “engaged in the business of selling such a product.” Id. §402A(1)(a). The same “business of selling” language is also found in Restatement (Third) of Torts, Products Liability §1 (1998). It may well be that hospitals, even if they have a few 3D printers, will still be considered not to be in “the business” of manufacturing medical devices with them. The same service/sale and public policy issues that drive the general rule that hospitals aren’t subject to strict liability would still be there. Lawyers for hospitals might want to bone up on casual seller precedent under the Restatements.
If the hospital that owns (directly or through some related corporate entity) the 3D printer isn’t the “manufacturer,” then who is? There would seem to be two choices – either the manufacturer of the 3D printer becomes the responsible “manufacturer” of literally everything that the printer can create, or the owner/licensor/whatever of the software itself is considered the “manufacturer.” The first result seems both unprecedented and harsh. There are literally zero product liability cases involving 3D printing at present (yes, we checked “trial orders,” too) at the present time. Imposing strict liability on the printer manufacturer for any product the printer could be configured to produce would be like imposing crashworthiness liability on the makers of the industrial robots used to make automobiles, or more generally on makers of a welding torches or plastic extrusion molders for everything that such equipment might be used to make. That’s hugely excessive liability and might well make 3D printing uneconomical.
The other option – the software designer – also generates controversy. Product designers, not being traditional manufacturers, have not been held strictly liable. This issue was litigated in breast implant litigation, where the original developer/inventor of the product (that had sold its relevant assets) was sued by plaintiffs afraid that the actual manufacturers might go bankrupt. The result was no strict liability. Christian v. Minnesota Mining & Manufacturing Co., 126 F. Supp.2d 951, 958 (D. Md. 2001) (developer of product, that no longer manufactured it, not strictly liable); McConkey v. McGhan Medical Corp., 144 F. Supp.2d 958, 963-965 (E.D. Tenn. 2000 (same, under slightly different theories); Barbour v. Dow Corning Corp., 2002 WL 983346, at *2-3 (Conn. Super. April 19, 2002) (same). In In re Minnesota Breast Implant Litigation, 36 F. Supp.2d 863 (D. Minn. 1998) (applying Arizona law), the court found no liability under any theory. First, that the original developer could not be strictly liable. “[I]f [the developer] played no role in the manufacture or the sale of Plaintiffs’ breast implants, [it]cannot be strictly liable.” Id. at 873. As for negligence:
There is no evidence to suggest that [the developer] was in the business of providing designs for breast implants. Admittedly, [it] designed implants for its own use. But [it] never sold breast implant designs to other companies, nor does Plaintiffs allege that [it] designed the implants specifically for Plaintiffs. Moreover, while [the developer] may have designed implants [earlier], once it sold the breast implant division . . ., it no longer had any control over the design. . . . Accordingly, the Court holds that [the developer] cannot be liable for negligent design with respect to Plaintiffs’ implants because [another entity] designed, manufactured, and sold the implants.
Id. at 873-74.
Accord Healy v. McGhan Medical Corp., 2001 WL 717110, at *3-4 (Mass. Super. March 29, 2001) (rejecting both strict liability and negligence theories against inventor that was not the product’s manufacturer.
We note that the allegations that weren’t present in Breast Implant are precisely those that would be present in a 3D printing case: the developer of the software used to make the device presumably was in the business of selling software, and the resultant medical device was designed “specifically” for the person who received it. Thus, as to the 3D software developer, precedent suggests that there might be negligence liability.
Other inventors, patents holders, and similar entities that had input into design, but not manufacturing, of medical devices have been held liable, if at all, solely in negligence. A defendant in Reeves v. AcroMed Corp., 103 F.3d 442 (5th Cir. 1997), was both the inventor of the product and chairman of the manufacturer’s board, but still could not be strictly liable. Id. at 449. In Murphy v. Aventis Pasteur, Inc., 270 F. Supp.2d 1368 (N.D. Ga. 2003) (applying Georgia law), there was no liability:
[T]he plaintiff alleges the existence of an affirmative duty to warn which arises merely from being the developer, inventor, or patent holder of a product or design. The plaintiff reasons that [defendant], as the holder of the patent on [the product], knew that other companies would copy [it] when the patent expired. The plaintiff further reasons that such knowledge on the part of [defendant] gave rise to an ongoing duty to warn the purchasers and recipients of such copied products manufactured by other companies. The plaintiff has not cited and the court has not found any authority supporting the existence of such a duty. Accordingly, the court concludes that no such duty exists. . . . Thus, the plaintiff’s negligence claim against [defendant] is not cognizable.
Id. at 1377.
In Kallman v. Aronchick, 981 F. Supp.2d 372, 382 & n.12 (E.D. Pa. 2013), under New Jersey law, “[t]he plain language of the NJPLA provides for no cause of action against an inventor, royalty owner, patent holder, or promoter of a product,” and under Pennsylvania law no “universal duty of care” exists “on the part of an inventor, patent holder, or royalty owner of an FDA-approved pharmaceutical product. Id. at 382 & n.12 (granting fraudulent joinder dismissal).
These are not necessarily the only analogies. Undoubtedly other categories of cases exist that separate design from manufacture. Liability claims against architects and other design professionals come to mind. So do cases involving allegedly wrong information in books, other publications, and software – although those cases often concern what is a “product” rather than who is the “manufacturer.” However, in the absence of any prior precedent, cases involving inventors and prior designers seem the most analogous.
The main difference, and where we get into what may well be uncharted ground, is that in all these other cases there was in fact another entity that obviously was a manufacturer. In the 3D printing situation, it is entirely possible that this technology has created a situation where there is no “manufacturer” at all within the meaning required by the Restatements for strict liability. The software manufacturer can be viewed as a designer, or maybe an author. The 3D printer manufacturer only made a tool, not a finished product sold to the plaintiff, and didn’t sell anything to the plaintiff. The hospital owning the printer is a service provider and also, at best, a casual or incidental seller, even where not expressly protected by additional malpractice-related statutes that vary widely from state to state.
It will also be interesting to see how the FDA chooses to regulate 3D printing of medical devices going forward. Whenever the FDA gets into the picture, preemption becomes an issue. A lot of these custom-made devices (they may also be “custom devices” under some FDA regulations, but we’ll just note that possibility and move on) are pretty complicated, created in a novel method, and have significant risks. Check out the slide show (unfortunately no longer available as of 2019). If they ultimately require pre-market approval, then who is the regulated party? The software provider? The hospital? Someone else? Whoever that party is, once the FDA requires it to obtain PMA approval for either the process or the finished device – that entity will have a preemption defense.
In any event, for the moment product liability litigation involving 3D printed medical devices remain (as far as we know) a gleam in the eye of the plaintiffs’ bar. But sooner or later, a 3D printed device will break, otherwise cause injury, or knowing plaintiffs, simply be in the general area when an injury happens. For every new cause of action, there are necessarily new defenses. It therefore behooves us as defense attorneys to start thinking about them now, rather than wait until we’re caught up in the heat of a novel battle.