How should manufacturers complete validation in order to be in compliance with current standards? In this report, two experts answer some of the most frequently asked questions involving medical device testing standards and how they affect the process of validation.
John Broad, SM, RM, NRM, is a senior consultant at NAMSA’s California laboratory and a specialist microbiologist with the American Society for Microbiology. He holds certifications in industrial sterilization and mycobacteriology and is affiliated with the American Society of Quality Control and the American Society of Microbiology. He also is active in AAMI/ISO sterilization subcommittees. David Parente, senior consultant and manager of NAMSA’s advisory services, has eight years of clinical experience in chemistry, bacteriology, mycology, parasitology, and immunology as well as more than 21 combined years of experience in process validation, contamination control, sterility assurance, GMP programs, biocompatibility testing, chemical engineering, and the quality control of medical devices. He also is involved with the writing and review of ISO TC 198 sterilization and packaging standards. NAMSA, headquartered at 6750 Wales Rd., Northwood, OH 43619, has laboratories in the U.S. and France, offers testing services, and manufactures sterility assurance products. U.S. labs are in Ohio, Georgia, and California. Broad can be reached at 949-699-6212 or and Parente can be reached at 678-449-0612 or

By John Broad and David Parente
When choosing a sterilization process for a medical device, U.S. manufacturers rely on guidance from FDA and validation methods developed by the Association for the Advancement of Medical Instrumentation (AAMI) as well as international standards. AAMI has been working with the International Organization for Standardization (ISO) and the American Standards Institute (ANSI) on the development of harmonized guidance for sterilization validation and testing methods. There have been a number of changes in the last decade, which have raised questions as to how manufacturers should complete validation in order to be in compliance with various standards. The following article examines frequently asked questions.
Guidelines and Standards
Q: What guidelines should be followed in order for sterilized medical devices to be compliant and acceptable for the international marketplace?
A: In order to harmonize guidelines in various countries outside the U.S., ISO has developed standards for commonly used sterilization methods. An example is radiation sterilization of medical devices. The standard is titled ANSI/AAMI/ISO 11137 Sterilization of Health Products–Requirements for Validation and Routine Control–Radiation Sterilization. The techniques developed by previous guidelines, AAMI ST31 and ST32, were adopted as ANSI/AAMI/ISO 11137 in 1994. These methods have also been referenced in European Standard EN 552. The European standard was prepared by technical committee CENTC 204 and has been given the status of a national standard implemented by Austria, Belgium, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Luxembourg, Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, and the UK. Concurrently, international standards for bioburden enumeration (ISO 11737-1) and sterility (ISO 11737-2) tests performed during validation processes have been accepted. The applicable European documents developed include EN 1174-1, which defines general requirements to estimate population of micro-organisms on products. Three additional drafts of EN 1174-1 are also under development. These documents (1174-2, 1174-3, and 1174-4) provide the manufacturer with assistance for implementing 1174-1 by providing specifics on sampling methods, validation of test methods, and test methodology. Technical comparison of all these documents shows greater uniformity in the critical areas of testing, which includes incubation times, media selection, and temperature conditions used during validation and routine auditing of the estimation of bioburden and culturing of products. Other methods finalized in 1995 include EN 556, which summarizes general requirements for terminally sterilized devices labeled “sterile” and the AAMI TIR 27 substantiation of 25 kGy as a means for establishing a 25 kGy sterilization dose for small batches. In addition, other means of sterilization guidance have been developed for various methods of sterilization. Table l shows standards used by manufacturers for radiation, ethylene oxide (EtO), and moist heat.

Q: What is required to sell a medical device sterilized by irradiation?
A: Section 820.75 of the Quality System regulation requires that all processes used to produce medical devices be validated, and this includes any sterilization process. Manufacturers have the choice of following accepted sterilization validation guidelines, such as those published by AAMI/ISO, or developing their own. Since the results of a sterilization process cannot be fully verified by subsequent testing, due to the destructive nature of the sterility test, the use of sterility testing is not recommended.
Sterility Assurance
Q: Can the USP sterility test be used to release product sterilized by radiation, EtO, or steam?
A: A manufacturer is free to select this test as a basis for release, but the radiation sterilization process must first be validated. After validation has been completed using accepted techniques, such as those found in the ANSI/AAMI/ISO guideline, dosimetric release can be used. Dosimetri or parametric release is a form of product release after sterilization that is based on the proper clearance rather than on a sterility test.

Q: What is an SAL? And what is the difference between an SAL of 10-3 and 10-6?
A: The sterility assurance level–or SAL–of a product is a measure of the probability that one unit in a batch will remain non-sterile after being exposed to the sterilization dosage. An SAL of 10-3 means that there is a probability that one device in a thousand might be non-sterile, and 10-6 SAL means that one device in a million could be non-sterile. A manufacturer must select an SAL because it is an integral part of the sterilization validation. In many cases, the intended use of the device will dictate the need for a particular SAL. Some European countries recognize only 10-6 for a “sterile” label claim. Therefore, the appropriate SAL may be based on the regulatory requirements of the country where products will be distributed. The commonly accepted SAL for invasive medical devices is 10-6.
Validation Methodology
Q: Is 25 kGy an acceptable sterilizing dose for radiation sterilized devices requiring no testing?
A: No. In the past, 25 kGy may have been used to sterilize product without a thorough validation, but since the GMPs have been in place, FDA has required that any sterilization method undergo some form of acceptable validation. In its 1991 guideline, AAMI adopted the term kilogray to describe an absorbed dosage level: 10 kGy = 1 Mrad. According to the model population used to establish Method 1 dose-setting procedures, a 25 kGy dose will sterilize a device with a bioburden below approximately 1,000 micro-organisms at a 10-6 SAL. However, many devices have a bioburden higher than 1,000 colony-forming units (CFUs) and would therefore require doses higher than 25 kGy to achieve a 10-6 SAL. The term bioburden refers to the viable single or multicell micro-organisms on a product prior to sterilization that have the potential for forming visible colonies.

Q: What are the key steps to take when planning a sterilization validation?
A: The sterilization of medical devices is considered a manufacturing process and must be initially validated and periodically audited. There are various ways to validate a cycle and, depending on the process, there are well accepted AAMI/ISO guidelines for them. The most common methods of sterilization are radiation, EtO, and steam. The key guidance documents for these processes are ANSI/AAMI/ISO 11137, ANSI/AAMI/ISO 11135, and ANSI/AAMI/ISO 11134. There are advantages and disadvantages to each process depending on the stability of the materials used for the construction of the device. (Table 2 offers a comparison.) The key steps to consider when choosing a method are function, packaging, safety, bioburden, endotoxin as applicable, application of biological indicators (BIs) as applicable, validation of testing methods, and residual analysis as applicable. In some cases, the device and its use will determine the extent of testing for safety/biocompatibility, the use of endotoxin testing, and the use of BIs.

Q: There are two validation methods in ANSI/AAMI/ISO 11137 and one in AAMI 13409. What are the basic differences?
A: Method 1 and 2 of the AAMI guideline involve establishing a sterilizing dose using a statistical model based on resistance. Method 1 makes use of an assumed heterogeneous micro-organism resistance model, allowing the dose to be calculated based on the bioburden levels found on the devices. With Method 2, the dose is determined on the actual resistance of the bioburden. The substantiation of 25 kGy as a sterilization dose relies on the bioburden and uses a minimum dose of 25 kGy. Method 1 and substantiation of 25 kGy require bioburden for the initial dose setting while Method 2 does not. Although the three methods require different numbers of samples to complete the validation, they all specify a quarterly dose audit that includes a 10 sample bioburden and a sub-process sterility test to confirm the continued validity of the sterilization method.

Q: Are there common validation methods used for processes such as EtO and steam?
A: One of the most common methods used in each process is known as the Overkill Method or Method C. Before this method is applied, studies should be performed to determine the effect of the process on product design and packaging. The desired SAL should also be defined by the intended use of the device. The SAL may vary between 10-3 (topical devices) and 10-6 (blood contact invasive devices). This method assumes the bioburden is less resistant to the BI and can be reduced to a minimum of 6 logs following a half-cycle exposure. The inactivation of the BI within the device should occur at the “worst case” location. After the successful completion of the method, release is typically performed by testing BIs. Once again the material surfaces can have an effect on the sterilizability of the device. Another commonly used method of validation for EtO sterilized devices is accomplished by conducting a series of partial cycles following one of two methods: Method A or B. Method A employs exposing BIs to graded exposures and enumerating the BIs. The data is used to construct a survivor curve. Method B requires using sets of BIs exposed to graded exposures, and the BI is sterility tested to determine survivors. The data generated from these studies is used to calculate a D-value. The resulting data obtained from Method A or B is used to determine the minimum EtO gas exposure time required for the full sterilization time. The parametric release of product following a successful validation is based on the conformance of the physical processing parameters established during the validation.

Q: Our company would rather not spend the time and money needed to conduct a sterilization validation. Can’t we just seed BIs, which consist of either spore strips or spore suspensions, in each sterilization lot and perform sterility testing of the BIs for lot release?
A: The use of BIs for validation or release of product is not recommended for radiation sterilization because there are some organisms that have a greater resistance to radiation than Bacillus pumilus, which is the standard BI organism used for radiation. The use of calibrated dosimeters to confirm dose delivery can also be accomplished as well as the dosimetry release of product without sterility testing. However, for sterilization processes such as EtO and steam, the use of BIs is widely used. For EtO sterilization, the EtO resistant Bacillus atrophaeus (a.k.a Bacillus subtilis var niger) has been employed for performance qualification studies.
Additional Online Questions
Q: How are BIs used for various sterilization processes for validation studies?
A: For processes such as EtO, the resistant Bacillus atrophaeus (a.k.a. Bacillus subtilis) has been used for performance qualifications studies. As found in literature, e.g. Block S, (see #26 in the “References” section at the end of this article) the organism has been characterized for resistance on a variety of surfaces and found to demonstrate various levels of resistance. However, scientists have found that resistance may vary when process parameters vary. When properly used as a process challenge device, BIs fit the sterilization process selected. The spores of the challenge organism are either placed onto paper carriers or directly inoculated onto the device surface determined to be the most difficult site on the device to sterilize. Caution shall be taken when using inoculum. Inoculum is not recommended for use during validation. This may cause inconsistent resistance to sterilization.

Q: Our company has numerous product lines. Do we have to validate the sterilization process used for each one separately?
A: Usually not, although it depends on the types of devices. Product grouping can be used to set up categories of devices or components and thus reduce the amount of testing needed to validate. It is entirely possible that all the devices a manufacturer produces can be considered one product group as far as sterilization dose requirements are concerned, but this needs documented justification. Standards for developing and monitoring product families have been previously established for radiation sterilized products (AAMI/ISO 15843) and for EtO (TIR) in preparation by the AAMI radiation committee. A similar approach has been developed for devices sterilized by EtO–AAMI TIR 28. Product adoption and process equivalency for EtO sterilization has been a useful tool for grouping and adoptions of new products.

Q: Our devices are custom made in small lots that can change frequently. How can we validate the sterilization process for these devices?
A: One way to do this for radiation sterilized devices is to use a prototype–a test sample made of the same material that is as large as the largest device you’ve manufactured to date, or plan to manufacture, and that has been subjected to at least as much handling and processing as your other devices and will not impact the level of natural bioburden contamination on the device. The object is to represent the “worst case” situation so that all the custom devices can belong to the product group represented by this “dose-setter.” Once you’ve validated the process, you will have to keep tight control over new or altered devices so that you can prove none of them poses a greater sterilization challenge than the dose-setter. Depending on the diversity of the devices, an alternative is to validate using a mix of all the products or components you manufacture. Additionally, the guidance is offered for small batch releases following ISO 11137 Method I or AAMI TIR 27. Your testing lab should help you establish data that can support an appropriate validation system. In cases where other processes are used such as EtO and steam, a prototype may be designed except when the lumens and mated surfaces of the device should be compared with the original design for gas/steam penetration. Small batch release for EtO sterilized devices is also covered in microbiological aspects, AAMI TIR 16.

Q: We want to validate using radiation sterilization Method 1, but our devices are large and costly. Do we have to use the full number of samples (140+)?
A: Not necessarily. The 30 samples that will be tested for bioburden have to be representative of the actual packaged devices. However, these same 30 samples can sometimes be used for the bioburden recovery validation (exhaustive method). The 100 samples specified for the dose verification study can be 100 sample item portions (SIPs) of devices. For instance, 20 devices could be cut into 100 or more portions, depending on the uniformity of the bioburden distribution. In some cases, this may not be possible because a large device might consist of several components, each having varying levels of bioburden. Each situation is unique, but there are usually several ways to reduce the total number of devices needed for Method 1 while still fulfilling the sample requirements.

Q: It takes several weeks to assemble standard lots of our product. Will this cause a problem for completing a Method 1 validation?
A: A Method 1 validation calls for 10 samples from each of three lots for bioburden testing. Using your current definition of lot, it could take months to complete the bioburden testing and several more weeks to complete the dose verification. However, you may be able to adjust the definition for this study. The purpose of using three lots for bioburden testing is to avoid basing the validation on limited data from just one set of production circumstances. If a lot can be redefined in such a way as to obtain samples representing a broad range of production factors, then this special definition can be assigned for the validation. For instance, a given day can have different production shifts, with materials or packaging from different lots, or any number of other meaningful definitions. A critical point to consider is to minimize the time that the bioburden samples are allowed to sit prior to delivery to the laboratory for testing. This could result in an artificially low bioburden that could affect dose calculation and the sterility tests.

Q: Should products that were validated using Method 13409 be revalidated using the new AAMI TIR 27 substantiation guideline?
A: It would be advisable to review the data from your Method 3 validation in light of the requirements for AAMI 13409 Substantiation Method. Rather than invalidating what was done, you may be able to simply “adopt” the Substantiation Method audit procedure based on the testing that has been performed. Caution should be exercised if the manufacturer is sampling below the imposed limit of 10. There is increased probability of failure with products with low bioburden in combination with small sample sizes. For example, products composed of stainless/titanium are manufactured under harsh conditions that will reduce the microbial populations to low levels. That population now may have a resistance probability model for the inactivation of the population. The manufacturer should consider increasing the batch size wherever possible to avoid validation failures. If revalidation is required, you should be able to continue sterilization based on the Method 3 data until the revalidation process is complete. Your testing lab should be able to advise you on the most practical way to ensure that current validation is met.
1. Sterilization of Health Care Products - Requirements for Validation and Routine Control - Radiation Sterilization, ANSI/AAMI/ISO 11137: 1994, Arlington, VA, Association for the Advancement of Medical Instrumentation, 1994.

2. Sterilization of Health Care Products - Requirements for Validation and Routine Control - Industrial Moist Heat Sterilization, 2ed. ANSI/AAMI/ISO 11134: 1993.

3. Medical Devices - Validation and Routine Control of Ethylene Oxide Sterilization, 3ed. ANSI/AAMI/ISO 11135: 1994.

4. Guideline for Gamma Radiation Sterilization, AAMI ST32-1191, Arlington, VA, Association for the Advancement of Medical Instrumentation, 1991.

5. 21 CFR 820.100.

6. Microbiological Methods for Gamma Irradiation Sterilization of Medical Devices, AAMI Technical Information Report, Arlington, VA, Association for the Advancement of Medical Instrumentation, 1991.

7. Sterilization of Medical Devices - Microbiological Methods - Part 1: Estimation of Population of Micro-organisms on Products, ANSI/AAMI/ISO 11737-1: 1995, Arlington, VA, Association for the Advancement of Medical Instrumentation, 1995.

8. Sterilization of Medical Devices - Microbiological Methods - Part 2: Tests of Sterility Performed in the Validation of a Sterilization Process, ANSI/AAMI/ISO 11737-2: 1997, Arlington, VA, Association for the Advancement of Medical Instrumentation, 1997.

9. Sterilization of Medical Devices - Requirements for Terminally Sterilized Devices to be Labeled “Sterile,” BS EN 556:195, Brussels European Committee for Standardization, 1994.

10. Sterilization of Health Care Products - Radiation Sterilization - Product Families and Sampling Plans for Verification Dose Experiments and Sterilization Dose Audits, and Frequency of Sterilization Dose Audits, 1ed.

11. Sterilization of Healthcare Products - Radiation Sterilization - Selection of a Sterilization Dose for a Single Production Batch, ISO 15844 Technical Report.

12. Sterilization of Single-Use Devices Incorporating Materials of Animal Origin - Validation and Routine Control Sterilization by Liquid Chemical Sterilants, 1ed. ANSI/AAMI/ISO 14160: 2000.

13. Chemical Sterilants and High Level Disinfectants: A Guide to Selection and Use, 2ed. AAM TIR7: 1999.

14. Principles of Industrial Moist Heat Sterilization, 1ed. AAMI TIR 13.

15. Contract Sterilization for Ethylene Oxide, 1ed. AAMI TIR 14: 1997.

16. Ethylene Oxide Sterilization Equipment, Process Considerations, and Pertinent Calculations, 1ed. AAMI TIR 15: 1997.

17. Process Development and Performance Qualification for Ethylene Oxide Sterilization - Microbiological Aspects, 1ed. AAMI TIR 16: 2000.

18. Radiation Sterilization - Material Qualification, 1ed. AAMI TIR 17: 1997.

19. Guidance for ANSI/AAMI/ISO 10993-7: 1995, Biological Evaluation of Medical Devices - Part 7; Ethylene Oxide Sterilization Residuals, 1ed. and Amendment. AAMI TIR 19: 1998; TIR 19/A1: 1999.

20. Parametric Release for Ethylene Oxide Sterilization, 1ed. AAMI TIR 20: 2001.

21. Sterilization of Healthcare Products - Radiation Sterilization - Substantiation of 25 kGy as a Sterilization Dose - Method VD max, 1ed. AAMI TIR 27.

22. Product Adoption and Process Equivalency for Ethylene Oxide Sterilization, 1ed. AAMI TIR 28: 2001.

23. Sterilization of Medical Devices - Validation and Routine Control of Sterilization by Irradiation, BS EN 552: 1994, Brussels European Committee for Standardization, 1994.

24. Sterilization of Medical Devices - Estimation of the Population of Micro-organisms on Product - Part 1, BS EN 1174-1 Requirements.

25. Sterilization of Medical Devices - Estimation of the Population of Micro-organisms on Product - Parts 2-4, Draft BS EN 1174-2 - Guidance on Part, BS EN 1174-3 - Guide to Methods of Validation of Microbiological Techniques, and BS EN 1174-4.

26. Block S, Disinfection, Sterilization, and Preservation, Fourth Edition, Lea & Febiger, Philadelphia, 1991.

27. Herring CM and Owens WM, “Experiences with the AAMI Dose-Setting Methods for Gamma Sterilization,” Med Dev Diag Indust, 6(6):50-55, 1984.


For additional information on the technologies discussed in this article, see Medical Design Technology online at or NAMSA at