General Aging Theory and Simplified Protocol

for Accelerated Aging of Medical Devices

Posted in Quality Assurance by mddiadmin on July 1, 1998
Share

The introduction of new or modified products to the medical marketplace requires the assurance that they
can be stored for an extended period (from one to five years) without any decrease in performance that
may affect safety and efficacy when the products are used. Because full-period, ambient-aged samples
usually do not exist for a such products, it is generally necessary to conduct accelerated-aging tests to
provide experimental data in support of performance and shelf-life claims for these products until fullperiod samples become available.
The ability of product designers to accurately predict changes in polymer properties is of critical
importance to the medical device industry. Modeling the kinetics of polymer deterioration is difficult and
complex, and the difficulty is compounded by the fact that a single-rate expression of degradation or
change developed over the short term may not be valid over the long-term service life of the product or
material being studied. In order to design a test plan that accurately models the time-correlated
degradation of medical polymers, it is necessary to possess an in-depth knowledge of the material
composition and structure, end-product use, assembly and sterilization process effects, failure-mode
mechanisms, and storage conditions.
Accurate prediction of medical product shelf-life performance is
critical. Photo: GATX Logistics, Inc.
A given polymer may have many functional chemical groups
organized in diverse ways (crystalline, glass, amorphous, etc.), along
with additives such as antioxidants, inorganic fillers, plasticizers,
colorants, and processing aids. It is the sum of these variations
combined with variations in product use and storage environment
that determines the degradation chemistry. Fortunately, the majority of
medical products are constructed from a limited number of polymers that have been well-characterized
over extended-use periods. A procedure known as the Simplified Protocol for Accelerated Aging (also
called the "10-degree rule") was developed around the collision theorybased Arrhenius model. When
applied to well-characterized polymer systems over moderate temperature ranges, the test results
obtained can be within the required degree of accuracy.
The aging of products or materials refers to the variation of their properties over time, the properties of
interest being those related to safety and efficacy. Accelerated aging can be defined as a procedure that
seeks to determine the response of a device or material under normal-usage conditions over a relatively
long time, by subjecting the product for a much shorter time to stresses that are more severe or more
frequently applied than normal environmental or operational stresses.
The primary reason for using accelerated-aging techniques in the qualification testing of a medical device
is to bring the product to market at the earliest possible time. The goal is to benefit both the patientfor
example, through early availability of a life-enhancing deviceand the companyby generating
additional sales and market sharewithout exposing either to any undue risk. Although accelerated-aging
techniques are well documented in academic circles, information on the use of these techniques in
medical product testing is somewhat limited. FDA has issued a handful of product guidances (for contact
lenses, human drugs and biologics, etc.) that incorporate accelerated-aging methodology, but no official,
broad-form agency policy currently exists. As a result, many medical product manufactures have

developed custom accelerated-aging policies based on these guidances or other applicable referenced
publications. (See the bibliography at the end of this article.)
Many accelerated-aging techniques used for the qualification testing of polymer medical devices are
based on the assumption of zero-, first-, and pseudo-first-order chemical reactions following the Arrhenius
reaction rate function. This functionlong the basis for studying most chemical reactionsstates that an
increase or decrease in the reaction rate at which a chemical reaction proceeds changes according to the
following equation:

where r = the rate at which the reaction proceeds; A = the constant for the material (frequency
factor);
= apparent activation energy (eV); k = Boltzmann's constant (0.8617 x 104 eV/K); and T =
absolute temperature. With appropriate substitutions, the simplified expression for the 10-degree rule can
be derived:

It should be noted that the 10-degree rule provides a conservative acceleration factor at room
temperature for activation energies less than 0.7 eV. Because of the exponential effect, this can be
conservative by orders of magnitude. (In some cases, a better match between the model and
experimental room-temperature data can be achieved by modifying the 10-degree rule using an alternate
temperature differential between 5 and 20C that best fits the experimental data.)
The 10-degree rule will likely be conservative in the prediction of shelf life. However, the technique
depends on numerous assumptions that must be verified by real-time validation testing conducted at
room temperature for the targeted shelf-life. A well-designed product-release test program will involve the
use of continued "room-temperature" aging that is always greater in age than the age of any product in
use. This is especially important when using these techniques for the qualification of critical (life-saving)
components or devices. The approach does involve some limited risk of potential recall, in the event that
room-temperature-aged testing shows a significant deficiency following real-time-aged testing of the
product. Applying accelerated-aging test techniques in conjunction with a comprehensive knowledge of
the materials involved is a prudent method of doing business, with the benefits of early product
introduction far outweighing the minimal risk of premature product failure.

Figure 1. Accelerated aging of polymers (time

versus temperature), showing the time (in
weeks) equivalent to 1 year of roomtemperature aging when a polymer is heat-aged
at a selected temperature (C). (Q10+ 10C
reaction-rate constant, assuming a room
temperature of 22C.)
The functional properties of products depend on
the properties of their constituent materials.
Since these materials for the most part are
polymeric in nature, their performance is related
to the rate of degradation of their inherent
structure and configuration over time. This
degradation is primarily chemicala
combination of effects arising from oxidative chain scission, oxidation hydrolysis, changes in crystallinity,
and other factors that may be environmentally dependent. Any stress factor that can reasonably affect the

functional properties of the product over time should be included in the accelerated testing program. The
testing is conducted at higher than usual levels of stresswhether of temperature, humidity, radiation,
temperature cycling, chemical environment, or other factors. The results at these accelerated stress
conditions are then correlated to those at normal operating or storage conditions using a physically
appropriate statistical model such as the 10-degree rule.

A number of miscellaneous factors must be considered in conducting an accelerated-aging study,

regardless of the accelerated-aging protocol employed:

When establishing the accelerated-aging protocol, the environmental conditions selected should
not represent unrealistic failure conditions that would never occur under real-time, ambient-aged
conditions. For example, where there is evidence that an aging effect occurs only in the presence
of heat, one should perform aging under conditions of storage or use only.

Proof testing can serve as a substitute for destructive testing. Proof testing requires all samples to
be tested at the end of each test interval to a specified failure point, then returned to the aging
environment for continued exposure. This method, however, is only applicable when the selection
of proof-test values does not weaken or compromise the product properties being examined.

When use of any accelerated-testing model produces a nonlinear plot, this may be an indication
that multiple chemical reactions, complex reactions of a second or third order, or autocatalytic
reactions are occurring at some, but not all, test temperatures. In these cases, elevated
temperatures may negatively distort the performance of the material at operating or storage
conditions. Under such circumstances, one should consider performing aging under conditions of
storage or use (ambient).

When possible, accelerated testing should be employedtesting of the device or material at high
stress for a short period of time in order to deduce the dominant failure mode. Based on
knowledge of the principal degradation mechanisms and stresses on or within the part, a
significant enhancement in test-plan efficiency can be achieved through the use of excessive
environmental stresses such as heat, oxygen, chemicals, or radiation. Often, radiation is the best
stressor to use, since the degradation pathways are often similar to those induced by heat or
oxygen. Irradiating a product at 100 kGy (10 Mrd) or more before initiating a formal test program
can often root out the product's "Achilles' heel" and allow for improved targeted test design. For
products currently on the market, additional insight into the most probable modes of failure can be
obtained from investigating customer complaint files.

All test units should consist of products constructed of the same components and subassemblies
and manufactured by the same processes, methods, and procedures as those used for routine
production. In addition, the productin its final packageshould be subjected to a minimum of

one standard sterilization cycle. Additional sterilization cycles or combinations of different

sterilization methods can be employed as needed to represent worst-case routine production.

For previously characterized polymers, a simplified approach for accelerated aging is based on
conducting testing at a single accelerated temperature and then employing the rule stating that the rate of
a chemical reaction will increase by a factor Q10 for every 10C increase in temperature. The typical
relationship selected for commonly used medical polymers is Q 10 = 2that is, a doubling of the reaction
rate for each 10C increase in the temperature above the use or storage temperature. The simplified
protocol for accelerated shelf-life testing is not a replacement for a more complex and advanced
accelerated-aging protocol, but is instead a protocol for systems known to conform to zero, first-order
Arrhenius behavior. This type of conservative relationship is appropriate for a wide range of medical
polymers that have been previously characterized.
Presented below is a list of the steps to be followed when designing an accelerated-aging test protocol
using the simplified 10-degree-rule methodology:
1. Identify the ingredients in the polymer formulation of interest. Care should be taken to identifyboth
qualitatively and quantitativelyall additives (e.g., antioxidants), fillers, and processing agents. Combine
this information with a thorough knowledge of the stresses on and within the part, identify the principal
degradation mechanisms for the system, and select experimental techniques that accurately evaluate the
degree of degradation and its impact on the performance of the product.
2. Select the reaction rate coefficient of Q10 = 2, unless another rate coefficient has been previously
determined experimentally.
3. Select an ambient temperature representative of actual product storage and use conditions (normally
between 20 and 25C). A temperature of 22C (72F) is preferred for most disposable medical products,
although any temperature that can be justified may be used.
4. In order to decrease the test time to the maximum extent, select a temperature for the acceleratedaging oven conditioning that is as high as possible, within the following limitations: (a) The acceleratedaging (oven) temperature must not exceed the material's glass-transition temperature (Tg), melt
temperature (Tm), crystalline-forming temperature (T), or heat-distortion temperature less 10C (T HDT
10C). This applies to all materials for which the above temperatures are greater than the ambient
temperature. (b) The accelerated-aging temperature chosen should not be greater than 60C, since the
accuracy of the assumptions on which this method is based declines sharply as greater geometric
multiplication factors are applied with greater deviation from ambient temperature. That is, any error will
also be multiplied by the same factor, resulting in a greater and greater error effect. (c) The preferred test
temperature of between 50 and 60C should be selected unless particularly temperature-sensitive
materials (such as PVC) are involved. At 60C, the accelerated-aging time relationship is that 3.7 weeks
is equivalent to 1 year ambient aging at 22C (room temperature).
5. For any accelerated aging and ambient temperatures selected, the relationship of oven test time to
shelf-life time is as follows:

where T1 = oven aging temperature, TRT = room temperature (ambient/ use/storage), and Q10 = reactionrate coefficient. As an example of the application of this formula, what test time in a 50C oven would be
required to achieve equivalency to 5 years of ambient shelf-life aging of a product at 22C (i.e., T 1 = 50C,
TRT = 22C, Q10 = 2)? The correct response is as follows:

In other words, an oven test time of 37.5 weeks at 50C would be equivalent to 5 years at 22C ambient
temperature (i.e., 7.5 weeks/year).
6. In cases where differences in coefficients of expansion in mating parts could contribute to significant
stress generation and part failure, temperatures need to be cycled through high (accelerated aging) and
then low temperatures. The high (accelerated aging) temperature is selected as described above in step
4, with the low (freezer) temperature < 5C. During the low-temperature conditioning, no test time is
accumulated toward the ultimate shelf-life equivalency of the product.
7. For certain polymer structures, long-term performance may be influenced by the effects of humidity
extremes. If humidity is to be included in the test design, a relative humidity > 85% for high humidity
conditions and < 20% for low-humidity conditions should be used.
8. Choose interim test intervals significant enough to detect the initial presence of failures, but not so
frequent as to invoke undue hardship on resources.
9. Select the specific properties to be evaluated, and the tests to accurately evaluate any change in these
properties.
10. Select a sufficient number of product test units so that statistically valid test results are obtained for
each test interval. The test units should normally be finished product; however, subassemblies and even
specially prepared test specimens are satisfactory in certain limited cases.
11. Test units drawn from the same product/material lot as the accelerated-aged units shall be ambient
(real-time) aged and tested according to the same test regimen and at the same test intervals as the
accelerated-aged test group. In addition, in order to control risk, ambient-aged product should be tested
for the ultimate shelf life as well as for any interim test period selected.
12. Following the steps listed above, develop a written test protocol specifying the accelerated-aging
conditions (temperature, humidity, heat cycling, time), time frames, sample sizes, and specific tests to be
undertaken at each test time interval. Note that because of multiple test groups (accelerated-aged,
ambient-aged, and control) and multiple test time intervals, sample sizes are relatively large and proper
resource planning needs to be executed to ensure adequate accelerated aging oven space, ambient
storage, personnel, and test equipment.
13. At the appropriate times, select samples from the aged and control groups and conduct testing as
specified.
14. Evaluate the product test results using appropriate statistical methods, developing required statistical
means, standard deviations, comparative t-tests, and F-tests to determine whether the product meets the
design specification criteria or control group comparison for each test interval.

Use of the simplified protocol for accelerated aging can represent a valuable means for device
manufacturers to obtain critical performance and shelf-life data on new products. There are, however, a
number of issues that should be kept in mind by anyone using this technique.
Results will generally be conservativethat is, lesser shelf life will be obtained compared with that
obtained via real-time aging. On the other hand, unrealistic negative outcomes may be produced because
of heat degradation. It is important to remember that the protocol is based on the assumption that all
materials in the study follow a zero- and first-order reaction-rate function, and that the supply of reactants
remains constant over the study time frame.

The process works best when the chemistry of the degradation reactions is well understood, and when
moderate aging temperatures are selected to minimize error factors and premature consumption of some
reactants (e.g., antioxidants, radical groups) that may be sensitive to elevated temperatures. One must be
alert for reaction-rate changes over test time frames; in such cases, use of multiple test temperatures or
alternate accelerated-aging methodology is recommended. Use of a different methodology may also be
advisable when a product is likely to include materials that have not been characterized previously. As
with any accelerated-aging method, there is a degree of risk until the study is validated with realtime/ambient testing. Finally, the testing regimen should be designed to provide data that are appropriate
to whatever criteria the product must ultimately satisfy.
A Review of Equipment Aging Theory and Technology, Electric Power Research Institute (EPRI) Report
NP-1558, Philadelphia, Franklin Research Center, September 1980.
Clark GS, Shelf Life of Medical Devices, FDA (DSMA) report, April 1991.
Donohue J, and Apostolou S, "Shelf-Life Prediction for Radiation-Sterilized Plastic Devices," Med Dev
Diag Indus,12(1):124129, 1990.
Gillen KT, and Clough RL, "Predictive Aging Results in Radiation Environments," Radiation & Physical
Chemistry,41(6): 803815.
Gillen KT, Clough RL, and Wise J, Extrapolating Accelerated Thermal-Aging Results: A Critical Look at
the Arrhenius Method, Albuquerque, NM, Sandia National Laboratories.
Meeker and Hahn, How to Plan an Accelerated Life TestSome Practical Guidelines, vol 10, Milwaukee,
WI, American Society for Quality Control, 1985.
Polymer MaterialsLong-Term Property Evaluations, Underwriters Laboratories Report UP 764B,
Northbrook, IL, Underwriters Laboratories, 1981.
Reich RR, Sharpe DC, and Anderson HD, "Accelerated Aging of Packaging: Consideration, Suggestions,
and Use in Expiration Date Verification," Med Dev Diag Indus, 10(3):3439, 1988.
Sandford C, and Woo L, "Shelf-Life Prediction of Radiation Sterilized Medical Devices," in Society of
Plastics Engineers, Inc., Technical Papers, vol XXXIII (ANTEC 87), Brookfield, CT, SPE, 1987.
Shelton WS, and Bright DG, "Using the Arrhenius Equation and Rate Expressions to Predict the LongTerm Behavior of Geosynthetic Polymers," Geosynthetics '93 (Vancouver, Canada), Roseville, MN, North
America Geosynthetics Society, 1993.
"Standard Practice for Heat Aging of Plastics without Load," ASTM Report D3045, West Conshohocken,
PA, ASTM.
Woo L, and Cheung W, "Importance of Physical Aging on Medical Device Design," in Society of Plastics
Engineers, Inc., Technical Papers, vol XXXIV (ANTEC 88), Brookfield, CT, SPE, 1988.
Karl J. Hemmerich is general manager and corporate technical advisor at Isomedix Corp.'s gamma
irradiation facility located in Sandy, UT. He was formerly president of Ageless Processing Technologies, a
consulting firm specializing in the medical disposables market, and has also worked at Ivac Corp., Cutter
Laboratories, and Becton Dickinson. He was a member of the task force that developed the technical
information report for postirradiation of materials (ISO 11137).