Medical Device Link.

Originally Published MDDI April 2003

Testing

The question of what to test is especially critical when the test report or results will be used to support a premarket notification, or 510(k), submission. In this case, it is important to keep in mind that the goal of the 510(k) is to demonstrate that the new device is substantially equivalent to one or more named predicate devices. This equivalence must be shown for all aspects of safety and effectiveness. 

All clinical features of the device should be tested, as well as safety, materials biocompatibility, environmental withstand, electromagnetic compatibility, and other areas, as applicable. (See Figure 1 for examples.) FDA guidance documents, where they exist for a particular device, are available on the Internet and can be very helpful. However, it is important to look at the issue date of a guidance document and to consider any possible changes in FDA thinking since then.

Figure 1. Examples of what to test for 510(k) clearance (Click to enlarge).

In selecting the tests to support a 510(k), it is necessary to look first at all claims made for the device in the indications for use, as well as in any advertising, marketing, or instructional materials. For each claim or feature, testing should show that the device performs satisfactorily across all reasonable ranges of patient and environmental conditions. The FDA reviewer will be asking many “What if…” questions when considering a 510(k) submission; the more of them that can be anticipated and answered in the initial submission, the faster the review will proceed. 

Years ago, the only way to demonstrate substantial equivalence was with side-by-side comparative testing of the new device and the predicate, showing that the new device was at least as good as the predicate. In 1998, FDA established the concept of recognized consensus standards. FDA publishes a list of U.S. and international standards that it recognizes as applicable to certain medical devices. The list is updated regularly. In brief, recognition of a standard means that FDA will accept testing of a new device against all applicable parts of that standard as an alternative to comparative testing against the predicate device. Of course, the new device must pass all of the tests. Testing against recognized standards could, obviously, save considerable time and cost.

FDA recognition of a standard can be complete, but exceptions do exist. The place to start is at the section of the CDRH Web site dedicated to standards, www.fda.gov/cdrh/stdsprog.html. That page provides links to more details of the application of recognized standards, as well as to the Recognized Consensus Standards Database, a search panel for specific standards. If a relevant standard is found, it is crucial to look at the page for that standard, which will list the date or edition of the standard, the devices to which the standard is applicable, and any exceptions or limitations to FDA’s recognition of the standard. Remember that the list of recognized standards is updated occasionally, with some being added and others withdrawn, as well as new editions of standards replacing old ones. It is also possible to formally request FDA recognition of a standard, but this process takes time.

When a recognized standard is not applicable, or when it doesn’t cover a particular aspect of a device, testing against the predicate is needed unless test data for the predicate can be found elsewhere (in promotional literature or prior 510(k)s, for example). It is often difficult or costly for the new product developer to obtain a competitive device—but there may be no alternative. In comparative testing, standards—even those unrecognized by FDA—can provide relevant test methods. Alternatively, it may be appropriate to develop new tests for this purpose.

Copyright ©2003 Medical Device & Diagnostic Industry