01.23.06
Get It Right Early and Fast: Testing and Analysis
As manufacturers place more emphasis on product development, contracting safety testing to outside sources is gaining popularity.
Frank Celia - Contributing Writer
A technician uses a materials tester to gauge the peel strength of a medical device package. Photo courtesy of North American Science Associates (NAMSA). |
But it is not going to be an easy ride. As demand and levels of technology increase, so does competition. Because testing and analysis firms are often at the ground floor of new product development, they are under additional pressure not just to produce accurate data, but to do so quickly. Experts predict that speed to market will be an ever more important factor in determining market share.
Moreover, orthopedic device technology is evolving faster than tests can be designed to evaluate it. Only in the last five years or so have hip and knee joint simulations become standardized (and some say the knee protocol is suspect). No dynamic simulation tests for shoulder or elbow implants exist yet, and those available for spinal implants are rudimentary. This is to say nothing of the monumental task of assessing new biologic innovations, such as plasma spray surface treatments or devices spiked with bone growth stimulation factors.
Following is a look at some of the experts (and their processes) who will face these challenges. Orthopedic manufacturers and the FDA look to these companies not just for accurate results, but, perhaps more important, to design, compare, modify and finally standardize the very testing protocols themselves.
Physical Testing
Building and testing prototypes are two of the most complex, time-consuming and expensive aspects of product development. Just about all the large orthopedic manufacturers have in-house facilities for dynamic and static physical testing, though few smaller ones do. However, even the large manufacturers can benefit at times from services offered by small, specialized testing companies.
The machines that perform dynamic testing are expensive, costing up to $400,000 (depending on the simulator’s complexity), and evaluating one implant can take between three and five months. Hence, manufacturers can easily develop temporary bottlenecks in testing new or modified products.
This Nelson pass-through sterilization chamber suite aids in the development and testing phase. Photo courtesy of Nelson Laboratories, Inc. |
Another service Li’s company offers is expert modifications on existing testing protocols. “As time goes on and we gain knowledge about how these devices fail, there is a need to continually modify the tests to mimic what we have learned clinically,” he said. Such modifications may involve making slight changes to existing simulators or building entirely new ones.
John P. McCloy, president of Cincinnati, OH-based Accutek Testing Laboratories, which also performs mechanical testing on medical devices, agreed that expertise and independence are qualities his clients value. More than half of his clients are orthopedic manufacturers, and much of what Accutek does for them is ensuring their products meet international ISO and American Society for Testing Materials (ASTM) regulations.
Another popular service offered by Accutek is the testing of products and components against one another to determine which is strongest, lasts the longest, etc. Obviously, a reputation for impartiality is paramount for this sort of task. “We are kind of like the referees at a football game,” McCloy explained. “We need to maintain our independence.”
One of the latest trends in physical testing is to avoid it altogether, at least in the early stages, and rely on data generated by computer simulation programs. In addition to being far less expensive than building physical simulators, computer simulation can save exponential amounts of time. In the early stages of a prototype, a solidly constructed computer simulation can save a manufacturer the cost and time of two or three flawed design cycles.
Simulation data are generated by using ANSYS software in a process known as finite element analysis (FEA), according to Brad Stevens of Kx Simulation Technologies, a Cincinnati, OH-based firm that was launched six years ago to specialize in computer simulations of orthopedic implants. Again, the large device makers all have ANSYS software, but they sometimes seek help from outside programmers for projects that require high-end analysis.
Testing in a hood at Nelson Laboratories, Inc. Photo courtesy of Nelson Laboratories, Inc. |
Computer simulations have become so prevalent that, on rare occasions, the FDA will accept computer data without having them confirmed by physical tests. Orthopedic manufacturers usually still have to complete physical testing, but the FDA may approve results in certain instances—such as a minor change on a pre-existing device.
Despite potential savings in time and money—or perhaps because the programming is so new, and computer experts and mechanical engineers inhabit such different worlds—Stevens believes his field still has to make further headway before being completely accepted on the factory floor.
“Product development is just getting hugely important right now, and we are, in a sense, in a battle with those engineers who still want to cut a piece of metal and test it versus our method,” he said. It is often the managers at manufacturing companies, those with the keenest eye on budgets and schedules, who are the strongest proponents of computer simulation options, Stevens added.
In the future, computers may actually be able to simulate how an orthopedic device will behave in relation to living tissue. DePuy Orthopaedics of Warsaw, IN has been experimenting with an algorithm that predicts to what extent implants cause bone remodeling, the process by which living bone tissue gains or loses density as a response to varying loading conditions.
Microbiology
The other major area of the testing and analysis industry is oriented more around medical research and laboratory expertise. These professionals deal with issues such as in vivo and in vitro toxicology, carcinogenicity, sterilization, analytical chemistry and cytotoxicity as they relate to orthopedic devices.
Most orthopedic devices are implants designed to remain in the body for a patient’s lifetime; thus, there is great concern over the initial reaction in terms of potential infection or rejection, as well as long-term effects resulting from potential chemical leachables that may cause cancer or have other adverse consequences. Scientists try to anticipate these potential problems using animal models, in vitro tests and chemistry experiments. The process often requires a wide range of viewpoints and expertise.
“There is often a team approach to some of this risk assessment,” said Paul J. Upman, PhD, director of scientific affairs at NAMSA, a large testing company based in Northwood, OH. “We have quite a crew of scientists here—chemists, microbiologists, toxicologists and even veterinary surgeons.”
One of the most significant trends in the orthopedic field is the move away from animal models. “The trend is to do more testing by chemical means, computer modeling [and] in vitro assays, to get away from the use of animals,” Upman said. This development is even more widespread in Europe and the United Kingdom than it is in the United States.
Humane concerns are part of the reason, but the most significant driving force is that animal testing simply takes too long, and newer testing methods can do the job better and faster. “As international standards such as ISO 10993 are updated, more emphasis is being placed on chemical characterization of materials, literature searches of available data and then a risk assessment by a competent scientist,” Upman explained, as opposed to animal trials, which can take a year or longer. (See “Will Animal Testing Always Be Necessary?” above.)
Another major concern facing these laboratories is sterilization. Recently this concern has taken the form of greater attention to what is known as residual manufacturing materials (RMM). Officials at the FDA believe orthopedic implants are at particular risk for this problem, because their manufacture requires so much cutting, buffing, polishing and sanding of medal. During manufacture, very tiny particles of matter such as metal shavings may attach themselves to the device and can end up inside the patient, where they can cause serious illness or rejection.
“There is one reported case of this causing a death,” said John Bolinder, sales manger at Nelson Laboratories in Salt Lake City, UT. “The FDA is very concerned about this right now.” His company is part of an ASTM team working to develop a standardized method for determining RMM.
Other responsibilities for Nelson’s orthopedic clients include cleaning validations on reusable trays and device kits. Many manufacturers provide the instrumentation required to perform the implant surgery, and there is a need to ensure that the instructions for cleaning and sterilizing those instruments work the way they are supposed to. Laboratories can accomplish this by simulating the kind of soiling an instrument might undergo during surgery (exposure to blood, bacteria, other pathogens, etc.), then following the cleaning instructions and evaluating the end result.
Sterilization methods will also play a vital role in the emerging field of resorbable products, according to Gerry Whitbourne of STS Duotek in Henrietta, NY. “The reason is those products must be packaged and sterilized in an environment that will let them be sealed for a long time without moisture and at the proper temperature,” he said. Moisture and temperature are the factors in the human body that will cause the product to start to break down, and it is important to ensure the breakdown process does not prematurely begin before the device is implanted.
Testing the Tests and the Testers
A central challenge in this industry is simply determining which testing protocols are most effective. “In large part, the labs that really do the best are the ones that help in developing those testing methods, making sure they are actually valid and yield reproducible results,” said Bolinder of Nelson Laboratories. To do this, companies participate in “round robin” testing, in which they all follow the same testing protocol and compare data to make sure the test is reproducible. These round robin tests are often arranged by ASTM or similar trade organizations. If a protocol proves successful, the FDA may at some point make it a standard requirement.
Companies (such as Nelson) that engage in this type of protocol evaluation can stay ahead of the learning curve. “When the FDA starts requiring those tests, usually we are at the forefront of being able to offer those services,” Bolinder explained. “The FDA does not always have the expertise to develop these tests on their own, and they really have to rely on the industry, manufacturers and contract service providers to help them determine what is best.”
This process is becoming more arduous in a rapidly expanding orthopedic market. Devices are being implanted in new locations (eg, spinal disc replacement surrogating fusion procedures). Current devices are being modified by biologics, and patient expectations are rising.
Devices are now being designed to restore function rather than just make a repair. And underlying all of this is the constant pressure to work faster and faster. Whitbourne of STS Duotek noted, “A lot of what we are seeing—particularly in the smaller companies, because they have most of the innovative ideas—is they are working very hard to get through the process as quickly as they can. There is a real race going on with a lot of new implantable devices.”
The performance of the testing and analysis industry will likely have a great impact on the outcome of that race.