Validating test methods

ISO 11607, TIR 22, and ASTM all help manufacturers understand their package testing responsibilities. 

Any medical device package test that is used to support conformance with ANSI/AAMI/ISO 11607, Packaging for Terminally Sterilized Medical Devices—Part 1 and Part 2:2006, must be validated. While not a new concept, test method validation continues to challenge packaging professionals throughout the industry.

According to section 4.4.1 of ISO 11607, “all test methods used to show compliance with this part of ISO 11607 shall be validated and documented.” AAMI Technical Information Report (TIR) 22:2007, reiterates the point in 5.2.1: “All test methods must be validated.” ISO 11607 and TIR 22 then point users to Annex B’s list of “suitable test methods.”

So are the methods of Annex B all you need?

“If you follow ISO 11607, an FDA consensus document, and use the test methods in the annex, you don’t have to create your own rationale as to why another test is appropriate,” explains Jackie Daly Johnson, president of Beacon Converters (Saddle Brook, NJ) and cochair of the AAMI Sterilization Packaging Working Group, which developed TIR 22. “You don’t have to reinvent the wheel every time.”

Many of the methods listed in the annex have been recognized by FDA as consensus standards. But just because a method is listed in the annex or recognized by FDA does not mean that it will work for every medical device manufacturer and for every medical device package. Determining the appropriateness of a particular method—and validating that method for a given package and in a given lab—is the responsibility of each medical device manufacturer.

YOUR RISK, YOUR RESPONSIBILITY

It may seem contradictory, but FDA regulations do give medical device packagers some freedom. “The beauty of FDA [regulation] is that it is up to you to pick a position and justify it,” says Laura Bix, PhD, assistant professor, Michigan State University School of Packaging. “But you’ve got to be able to defend that position. There is a degree of uncertainty, so the choices you make depend upon the risks you want to assume. If you cannot defend your position, it could mean that you are not correct.”

Companies, though, may look for a less risky course of action. And that course may keep them on a path well traveled.

“Companies are looking for something in black and white,” explains Ondrea Kassarjian, training programs manager for Lansmont Corp. (Monterey, CA). For some, black and white may mean following standards tested by peers and recognized by FDA.

Explains one of Bix’s graduate students, Raghav Prashant Sundar: “As a graduate student engaged in active research, I feel that it always helps to keep pace with ASTM and ISO standards. Doing so enables us to use standardized test methods in our research projects. On a personal note, I have always felt very comfortable while talking to someone about research projects when the test methods used are based on a standard. I always seem to have a strange sense of self confidence while presenting research findings or proposals when they are based on a standard rather than some in-house method.”

UNIVERSAL UTILITY?

Alas, there is safety in numbers. If a test method has been tested for reproducibility and repeatability in several different settings, and it proves to be so, the method can be standardized. But does that mean the method is more reliable and more likely to reveal your package’s potential defects? To answer that question, you must validate that you can perform and repeat the method in your laboratory accurately with the precision necessary to identify defects particular to your package.

“Only consensus methods with precision and bias [P&B] statements are cited in TIR 22, which makes those methods easier to validate,” says Johnson. “Users compare their own facility tests—their own lab’s variability—to the P&B statements published in the test methods.”

Explains TIR 22 in 5.2.1.1: “Test methods that have been subjected to rigorous inter-laboratory studies may be preferred since the repeatability and reproducibility have been determined. When incorporating these methods into a specific laboratory it is important to demonstrate the accuracy and repeatability of the method are at least as good as the reproducibility from inter-laboratory studies.”

Says Michael L. Troedel, president of Troedel & Associates Inc. (Lake Villa, IL). “I believe the publishing of ISO 11607 Parts 1 and 2, TIR 22, and FDA recognition of physical test methods for package integrity have all led to an increased awareness of the need to validate and test medical device packaging. Furthermore, most people are just now finally becoming aware of the need to document the repeatability and reproducibility of their test methods to ensure their data being generated is valid.”

TIR 22 does say that other tests may be used. Reads 5.2.1.2: “Test methods developed independently or from the scientific literature may be used. However it is important to determine if the test method meets the required sensitivity and that the accuracy and repeatability meet predetermined criteria.”

ASTM F2097, “Standard Guide for Design and Evaluation of Primary Flexible Packaging for Medical Products,” can serve as another source of support for medical device packagers. “It complements TIR 22, presenting things a different way, such as in charts. The similarity is not coincidental—the ASTM document was written first and was very useful as a compendium in writing the guidance in TIR 22,” says Johnson.

Hal Miller, principal of PACE Solutions LLC and cochair of ASTM Committee F02 (Flexible Barrier Packaging), says that the committee is preparing a compilation of flexible barrier packaging design standards that will provide the user with all the most commonly used standards referenced in F2097. “This guide references specific individual test methods, their descriptions, and their applicability for testing, research and development, or compliance based on the characteristics of the specific product to be packaged. The design and evaluation guide is organized into several categories for evaluating flexible barrier packages and packaging materials. Each category has specific material or package properties or functions to be evaluated. The organization of the compilation follows this same structure with the standards included.” The compilation will be published in both book and CD form, and the committee hopes to have it available in the spring.

Adds Troedel: “The compendium, as currently planned, will include medical device test methods from multiple sources. It will be all inclusive and cover topics such as safety properties, barrier properties, durability, package integrity and seal strength, visibility and appearance, processing properties, transportation simulation, guides, conditioning, and definitions.”

ASTM committees play a crucial role in easing decision-making in the industry. “Through ASTM, we try to create test methods that people can utilize,” says Steve Franks, executive vice president of T.M. Electronics. He reports that ASTM Committee F02 is working to establish a new method, designated WK19529, “Nondestructive Pressure or Vacuum Decay Leak Test for Sealed Products or Packages.” Says Franks: “An interlaboratory study should be underway in the next six months.”

Another new test method is ASTM F2714-08, “Standard Test Method for Oxygen Headspace Analysis of Packages Using Fluorescent Decay.” Mocon Inc. (Minneapolis) will be participating in round-robin tests to provide correlation data with the method, reports Ed Emerson, business development, Mocon.

READY, SET . . . VALIDATE

As “off-the-shelf” as the methods in Annex B of ISO 11607 seem to be, though, if a medical device manufacturer has chosen to use a listed method, there is still work to do. “I often hear a potential user say, ‘I will just use it,” when considering a method,” say Franks. “But the reality is that the method and equipment all need to be validated with a company’s own systems and with its own operators. They have to validate that the whole process of using the method is reliable in their lab and ultimately ensures that the package being tested meets its performance requirements.”

Concludes Franks: “The wonderful thing about packaging is that there are no two packages that are alike, so no two testing programs will be alike.”