Wider use stymied by regulations, materials, lack of knowledge
The use of additive manufacturing (AM) in the medical industry is well established in making dental implants, artificial hip joints, and molds for invisible braces. The layer-by-layer technique, also known as 3D printing, promises to grow further, aided by better design tools and increasing numbers of design engineers trained in AM, the possibility of faster prints, a flattening cost curve, automation, and the need for patient-specific devices that can’t be made as well or at all with traditional manufacturing.
Holding back wider use of AM are regulatory requirements, a limited set of materials, and lack of familiarity with its possibilities in the medical community.
Laura Gilmour, global medical business development manager at one of EOS of North America Inc.’s technology innovation centers (Pflugerville, TX), said, “In medical it is much [further] ahead than some other industries, and for medical device manufacturers they’re realizing that if you’re an orthopedic device manufacturer and you’re not looking at additive manufacturing, you’re behind.”
As some of the early adopters of AM, medical device makers still employ the method for prototypes, where it’s widely used, but have moved their output numbers upward into production-scale territory, Gilmour said.
“We have some customers producing several thousand hip cups per month,” she said. “That, to me, is at a production level. And then we have other customers creating patient-specific cutting guides at several hundred per month.”
Gilmour mentioned a plastic (cutting guides) and a metal (hip cup) example in her comment. Considering the two material classes separately is useful to a discussion of whether AM production costs can be reduced to compete with traditional manufacturing.
Cost and Performance
“This question comes up a lot across the industries that we serve,” said Bob Yancey, director of manufacturing industry strategy at Autodesk Inc. (San Francisco). “The cost curve on additive continues to come down. In most cases, it’s still not competitive with traditional manufacturing, but there are certain areas where it’s starting to come close.”
As an example of AM production costs nearing those of a traditional manufacturing method, Yancey mentioned plastic parts that in the past have been made using injection molding.
“Lots of disposable medical products are made with plastic injection molding,” he said. “So, you’re seeing some of the plastic additive manufacturing methods starting to compete with injection molding in lot sizes in the tens of thousands.
“On the plastic side, I’ve been hearing somewhere around 50,000 is the break-even point between additive and injection molding,” said Yancey. “I think that’s very application dependent so that’s probably a best-case scenario.”
Stratasys Direct Manufacturing (Los Angeles), the manufacturing arm of Stratasys Ltd., is nowhere near a run of 50,000 in its use of 3D-printed silicone tools to cast urethane housings for low-to-mid size production runs of medical devices. But the AM process offers other advantages in plastics manufacturing. At times, using the silicone tools and the nimble AM process can offer a first-to-market advantage or a bridge to start production while a metal tool for injection molding is being made.
Also made of plastic are polyamide, patient-matched surgical guides used in the operating room that help surgeons cut and drill into bone. “Traditionally manufacturing the guides required long lead times and was pretty expensive to do using traditional techniques (i.e., clay molding and casting into molds),” said Benjamin Johnson, director of product development for healthcare, 3D Systems Healthcare Inc. (Littleton, CO). “It was never really a viable process and was done for the most complex of cases where there weren’t really any other options.”
Due to the reduced operating room time, improved surgical outcomes, patient satisfaction, and surgeon and patient confidence, Johnson said, orthopedic surgeons have joined oral and craniofacial surgeons in using nylon cutting guides. This is helping to drive up their production numbers.
On the metals side, there has been almost a doubling of the number of metal additive manufacturing machines sold in the last year, according to “Wohlers Report 2018: 3D Printing and Additive Manufacturing State of the Industry.”
“So, there’s clearly a lot of growth in that industry and it’s primarily being driven by aerospace and medical,” Yancey said.
While the use of AM for metal devices is much smaller than it is for plastics, there are some new techniques that may have the potential to be competitive in lot sizes of hundreds to thousands, Yancey said.
“New entrants on the metal side—Desktop Metal and MarkForged—claim they can compete on cost with metal injection molding,” he said. “HP and GE have also announced technology that they say will compete with metal injection molding and small castings.
“Metal injection molding is rarely used for medical applications so I’m not sure what impact these new methods will have on the medical industry,” he continued. “[However], the new metal methods will be more cost-efficient than the methods used today, which are primarily direct metal laser sintering, which is based on selectively melting metal powder with a laser.”
One thing that is certain is the use of AM for features on metal implants that foster bone in-growth, which is good for long-term stability in the body. Prior to 3D printing, manufacturers used titanium plasma spray to create a roughened or porous and reticulated, bone-friendly, osteoconductive surface on an implanted device. For this application, plasma spray was less than ideal.
“Traditionally, plasma spray coating is a very expensive and high waste type of process,” Johnson said. “So, yields aren’t very good. But 3D printing allows you to exactly print the structure.”
Gilmour pointed out another advantage. “You get the added benefit that it’s one piece, without the risk of delamination with the use of a surface material with a binder to a substrate,” she said. It also eliminates the risk of contamination from a binder.
Gilmour pointed to a competitive advantage from 3D printing for both plastic and metal devices—being first to market and potentially being the leader in scale production. “If you’re first to market with a new innovation or device, you’re going to have a clinical history faster,” she said. “Surgeons are most comfortable when they know the 10-year clinical history of a technology.
“If you are the first to a successful innovation, you’re the first to get to 10 years of clinical history,” Gilmour continued. “Therefore, your next innovation using the same technology type could be even more widely accepted by the medical community.”
AM offers other cost and competitive advantages for both plastics and metals.
“The ability to create a near-net-shape part from a 3D printer typically reduces the amount of machining that needs to be done, and there’s cost savings in that,” said Johnson of 3D Systems Healthcare. “Complex geometries are easier and cheaper to produce using a 3D printer than they are to traditionally manufacture. In patient-matched devices, the ability for a 3D printer to print a structure that conforms to a patient’s anatomy is something that’s much more readily done and cheaper than it would be to cast or mill a part.”
A common complaint about AM is it’s slow. Companies such as HP and Carbon 3D have focused on that issue and have developed technology that promises faster printing. It may be some time before these machines benefit the medical industry, however, since they don’t yet use materials used in devices previously cleared by the U.S. Food & Drug Administration, according to Autodesk’s Yancey.
Gilmour added a final thought: “The cost of production [AM vs. traditional manufacturing] depends on what you’re making. You really need to be able to use some of the design features [of AM] to get the advantages. So, if you’re not open to changing your way of thinking, AM may not provide as much of an advantage.”
Making Better Designs
Sometimes, making a device with AM depends on the ability to redesign it. Generally, it’s not possible to take a design made for traditional manufacturing and apply it to AM. Designers quickly learn that AM has three advantages: creating complex designs, making something custom or patient-specific, and reducing the number of parts and incorporating functional integration to the design.
“The design of devices is completely changing because of additive,” said Johnson. “There’s a whole new set of rules that design engineers need to learn in order to take advantage of 3D printing.”
Johnson said opening the door to additional applications supported by 3D printing is dependent not only on the speed and scale of the equipment but also on materials and a shift in design paradigms.
“The more engineers learn about 3D printers and what can be done using them, [the more the] door will open to new applications,” he said.
For example, AM was used to print structures that perform more like physiologic tissues, including spine interbody implants that mimic the elastic modulus of surrounding bone and help reduce undesirable side effects after spine fusion surgeries.
“That’s an example of how the freedom in design for 3D printing allows you to create more complex structures that you could not traditionally manufacture that lead to a better outcome for the patient and a better performing device,” Johnson said.
Helping design engineers make better designs for AM is the 3MF (3D Manufacturing Format) Consortium (Wakefield, MA), which recently released its Beam Lattice Specification Extension to its 3MF Core Specification. This is 3MF’s fourth specification extension, and is a new method for storing and transferring lattice-type geometry information that’s useful in promoting bone in-growth into implants.
The goal of 3MF is to define an open source, universal AM file format that will allow design applications to send full-fidelity 3D models to a mix of other applications, platforms, services and printers and eliminate the widespread issues with currently available file formats. Coincidentally, the companies interviewed for this article—3D Systems, Autodesk, EOS, and Stratasys—are all founding members of 3MF, among others.
There are other developments intended to help a design engineer work in AM.
Autodesk’s Yancey said there are more automated tools being developed that allow a designer to use the data in a CT or MRI scan to create the geometry and fixturing required for a custom implant or prosthetic. Manual work is still required to separate out the region of interest from the CT or MRI data. This data comes in a 3D grayscale image and separating a bone or a tumor from other soft tissue from the surrounding data can be challenging because the grayscale variation may be small. It often requires someone with good computer skills and expertise in human anatomy, Yancey said.
Johnson agreed: “Traditionally, separating out anatomies is a painstaking, slice-by-slice manipulation of the imaging data.”
One solution is 3D Systems’ D2P (Dicom to Print) image processing software. With CT scans, which are more commonly used for bony structures than MRIs, it lets the designer separate out anatomy based on the Hounsfield unit or the radio-density of the tissue. Similarly, there are algorithms in D2P that allow a designer to pull in contrast enhancements for cardiovascular tissues and similar applications.
What’s Holding it Back?
With all of the advantages of AM—the ability to create complex designs that are impossible to make any other way; tailoring a device to a patient; and parts consolidation—what’s holding it back from more widespread use?
One issue may be the materials available for use. The FDA doesn’t approve materials, but getting the first device made of a certain plastic or metal takes much more effort and money than it does for the second or any subsequent device made of the same material. Partly for that reason, newly introduced materials are rare, and the primary metals used in AM for medical are titanium, cobalt-chrome and stainless steel. On the plastics side, it’s polyamide, nylon, acrylonitrile butadiene styrene (ABS), and polylactic acid (PLA).
The selection of materials is part of the regulatory process. SME’s inaugural “Medical Additive Manufacturing/3D Printing Annual Report 2018” sheds light on this and other issues (available for download at www.sme.org/medical-additive/). In the report, of 181 respondents from medical device manufacturers, hospitals, AM technology developers, universities, government agencies and more, 97% said they expect the use of AM to grow.
But, as a majority of those surveyed (61%) pointed out, the regulatory process is holding back broader use of AM in the medical industry. Other roadblocks include the need for more materials (47%); not enough capital (44%); gaps in a qualified workforce (41%); the need for better processes and printers (39%); reimbursement issues (31%); the need for better processing software (25%); competition (~11%); and other, unspecified issues.
In another example of a regulatory roadblock, consider that people undergoing hip replacements get an off-the-shelf implant that’s deemed the closest fit. So far, the FDA allows additively manufactured, patient-specific devices only for very specific cases, where a severe revision (e.g., cancer-affected bone) for the acetabular side of the hip must be done. These types of surgeries are typically done at university-affiliated hospitals, which may put them out of reach for people living in rural areas or for those without the means to access large academic medical centers.
“I do see that likely changing, but any regulatory change like that will take quite a bit of time,” said Autodesk’s Yancey.
Another impediment to broader use stems from the community of medical professionals and their lack of knowledge.
“The biggest impediment mentioned [at SME’s RAPID + TCT conference] is that most surgeons don’t understand what they can get out of the technology, so a lot of it’s just education and then having the resources in the hospital to produce the model and the expertise to process the CT or MRI data to be able to generate the model,” Yancey said.
Getting some doctors to embrace AM might be an easier task than with others.
“In the past, we had to sometimes tell a surgeon what he/she wanted wasn’t possible to make, but AM has changed that,” Gilmour said.