Page 240 - Biomedical Engineering and Design Handbook Volume 2, Applications

STERILE MEDICAL DEVICE PACKAGE DEVELOPMENT 219

So, in the example protocol, the accelerated aging study is shown as a parallel path to the envi-
ronmental challenge and distribution and handling segments of the flowchart. Other protocols hav-
ing the same elements may be designed to comply with the requirements of the standard.

7.9.1 Validating a Package “Family”
In order to reduce the burden of testing, thought should be given to other package system designs and
devices that may require validation. Similar medical devices and package designs may constitute a
“device family” and may be tested as a group when packaging materials, manufacturing/assembly
machinery, sterilization process, and all other aspects of the device and package system are compara-
ble. In this case the worst-case configuration for the device, materials, processes, and logistics for the
package must be validated. A rationale for determining the worst-case configuration must be devel-
oped to support the technical decision. Another consideration for reducing the burden of testing might
be to leverage existing package systems that have been validated in the past. Here the burden will be
to demonstrate and document the similarity between the package systems in question.

7.9.2 Worst-Case Package System
When the worst-case scenario is being used to cover multiple package systems and designs, a dis-
tinction must be made between the worst-case sterile barrier system and the worst-case package sys-
tem. This distinction was necessary because the revised standard placed additional emphasis on the
concept of testing the “worst-case” package configuration. But it was not clear what was meant by
this requirement. The AAMI Packaging Committee helps by clarifying this requirement by stating that
“Don’t confuse worst-case process parameters with worst-case package configuration. The require-
ments for a worst-case process is discussed in ISO 11607-2:2006 clauses 5.3 and 5.4. The worst-
case may not be (at) the lower limits. The use of worst-case SBS does not necessarily require
MDMs to acquire from sterile packaging manufacturers (SPMs) special lots of preformed SBS
made at worst-case conditions specifically for design performance qualification. It means that when
the MDM is placing a closure seal on a preformed SBS or is sealing a lid to a blister tray, for exam-
ple, he or she should seek out the process variable that produces the weakest SBS and test samples
made under those conditions. When addressing the design configuration worst-case (scenario), one
approach is to evaluate the worst-case configuration for the device and package family. Using IV
sets as an example, these products can range from simple tubes with fittings to a complex system
of tubes, ports, and valves. If you wish to classify IV sets as a family that can be packaged in, for
example, a specified header bag, you must identify the family, determine the worst case example
(the one with the most tubes, ports, and valves, for instance), and use it for the design qualification
performance testing.”

7.9.3 Test Sample Size Considerations
This is one of the most confounding aspects of developing the final package system validation plan.
The protocol plan must weigh the economics of providing a high level of confidence against the risk
associated with producing a package system that will fail and cause harm to the public. Here’s how
the SPMC answers the question of what sample size to use for package integrity testing:

“It is not possible to simply recommend and ‘appropriate’test quantity for attribute data (testing) whether
in initial, pivotal, or release stages. The subject of sample size determination is not a trivial one, and in
the medical device area is directly related to managing risk. Risk varies significantly with the nature of
the medical device, particularly for those designed for implantation. Also of significant concern is the
reliability and reproducibility of the measurement system, and the degree to which those systems have
been validated. These and other considerations are generally considered toward the end for determining
an appropriate AQL level, which can then be used to determine sample sizes for testing.”