Originally published February 1996
Paul J. Sordellini
After decades of anguishing over long aeration times and interminable incubation periods, manufacturers of medical devices sterilized with ethylene oxide (EtO) may finally be getting something of a break. Thanks to the efforts of industry experts in the United States and abroad, significant attention is now being paid to the possible use of parametric release for EtO-sterilized devices. Their work is providing industry with safe and complete guidance that may soon make parametric release an everyday reality. As a result, companies can expect to see a dramatic improvement in the turnaround times for products sterilized in-house or out sourced to contractors.
It is estimated that approximately 45% of the medical devices manufactured in the United States are sterilized using gamma irradiation. Another 45% of the market is claimed to be held by EtO, while the remaining 10% is held by steam and electron beam (E-beam).
When using any of these four sterilization methods, manufacturers must validate the process in order to provide documented evidence that it will consistently yield the desired sterility and sterility assurance level (SAL). Upon completion of all stages of validation, routine control procedures for gamma, E-beam, and steam consist of monitoring the physical parameters of the process. Products may be released as soon as it has been confirmed that the routine production cycle has fallen within the parameters established during validation. For gamma and E-beam those parameters are exposure time and absorbed dose; for steam, they are exposure time and temperature. In either case, the time required before products can be moved out of the poststerilization phase and shipped to market is generally dictated by the time necessary to perform a routine quality assurance review of the physical process parameters verified during the sterilization cycle. Release of products on this basis is called parametric release.
While parametric release is the norm for most sterilization methods, however, it has rarely been employed by contract EtO sterilizers. Following completion of an extensive validation process, EtO sterilizers have continued to use biological monitoring during routine production cycles as the basis for product release. The typical process used for EtO sterilization includes initial inventory control checks, placement of biological indicators (BIs), preconditioning, time inside the sterilization vessel, aeration, retrieval of BIs, shipment of BIs to the testing laboratory, BI preparation, as many as seven days of BI incubation, compilation of the test result report, and communication of test results to the manufacturer--who can only then arrange return shipment of the lot to the manufacturing site for inspection and distribution. In some cases the resident time of a given product lot within the walls of a contract EtO facility can be as long as 11 days, and it can be even greater if weekends intervene to hinder BI preparation or retrieval.
Just as the practice of parametric release by EtO contractors has been rare, so has been the availability of related guidance documentation. The 1988 standard compiled by the Association for the Advancement of Medical Instrumentation (ANSI/AAMI ST27) dedicated only one brief paragraph to the subject of "process control release," stating that if controls are sufficiently reliable, process control release may be considered.1 The absence of further detail-- together with the accompanying note that regulatory authorities such as FDA may have to grant their approval for process control release--undoubtedly prevented many EtO sterilizers from seeking to use this technique.
In 1994, U.S. delegates voted to adopt a newer standard compiled by the International Organization for Standardization (ANSI/ AAMI/ISO 11135) in place of ANSI/AAMI ST27.2 This document contains updated ISO-compatible requirements for validation, control, and conventional BI-based release of EtO-sterilized product. But it goes beyond earlier standards by including a full section on parametric release, which details what needs to be done differently during validation and routine control in order to release product parametrically. Annex D of the standard is also devoted to this subject and furnishes specifics regarding management expertise, temperature spread across the load, monitoring of effective recirculation, and performance qualification. In the hands of a sterilization expert, ISO 11135 offers enough information for a contract sterilizer to begin instituting parametric release and, in fact, industry interest in this time-saving system has increased dramatically.
THE POTENTIAL OF PARAMETRIC RELEASE
The use of EtO for sterilization is as old as the equipment commonly employed. Parametric release of devices processed with gamma radiation, E-beam, and steam is also an old practice. However, because EtO sterilization requires accurate control of a greater number of variables--thus making it somewhat of an art rather than a pure science--industry has preferred to continue using BIs to confirm the successful integration of all the critical parameters. In this context, the routine use of parametric release would represent a significant advance in the field of industrial EtO sterilization for medical devices.
The present push toward making parametric release a routine part of the world of EtO sterilization is the result of two forces. First, by adopting ANSI/AAMI/ISO 11135, industry finally has an internationally recognized document that lists the general requirements for parametric release. Second, to keep up with the trend of cost cutting that is affecting all sectors of the health-care marketplace, the medical device industry has a desperate need for greater efficiency in all its processes.
One response to the need to reduce health-care costs has been to eliminate massive stockpiles of medical products and switch to just-in-time (JIT) inventory systems. This changeover applies both to hospitals, which now order only what they need and only when they need it, and to manufacturing companies, which now regulate their production according to immediate market demand. To implement a JIT system, the manufacturing sector must upgrade its ability to respond efficiently to periods of increased market demand for its devices. For many device manufacturers this has meant switching to gamma, E-beam, or steam sterilization, where the opportunity to use parametric release translated into quicker response time and reduced costs. But for other manufacturers whose devices suffer from material incompatibility with those methods, EtO has remained the only viable sterilization option. And with it has come long turnaround times and limited flexibility in responding to market demands.
The technique of releasing EtO-sterilized product based on process parameter review has the potential to revolutionize the EtO industry. Once validation of the process has been completed, use of parametric release will streamline the EtO sterilization process to initial inventory control checks, preconditioning, sterilization, and aeration, followed by direct shipment to distribution sites. Allowing a few hours for initial and final inventory control, and 12 hours each for preconditioning, sterilization, and aeration, it is conceivable that a product lot could enter and exit an EtO sterilization facility in about 40 hours--thus rivaling the turnaround times offered by other sterilization methods. Other products may require more time, depending on their resistance to the process, vacuum sensitivity, and aeration requirements.
Added advantages of parametric release are less product handling and related damage. Many manufacturers have experience with the damage and delays associated with cutting open boxes to place BIs, reopening the same boxes after processing to retrieve the indicators, lost or incorrectly placed BIs, and laboratory errors in handling or incubating BIs. There is also the expense of purchasing, placing, retrieving, and testing the BIs to consider. In addition, once a contract sterilizer has handled a product lot so extensively, manufacturers usually require that the load be returned to the manufacturing site for a careful quality inspection.
By contrast, a load that is going to be released parametrically can be net wrapped by the manufacturer and processed without ever having a single carton removed or opened. Net wrapping permits temperature probes to be inserted between cartons at sites appropriate for monitoring validated temperature parameters. The load can then remain intact throughout the sterilization phase and be confidently shipped from the sterilization site directly to a distribution center--thus avoiding the time and expense of bringing it back to the manufacturer. In harmony with JIT, some manufacturers could have products EtO-sterilized and on their way to market less than 48 hours after they come off the assembly line.
To be sure, implementation of parametric release will require EtO sterilization firms to make substantial financial and professional commitments. The costs of installing and operating the additional gas analysis equipment required for parametric release can be considerable. Company management will require additional training, validation protocols will have to be reworked, and standard operating procedures will need to be upgraded. And, most important, the company will need to establish clear and efficient channels of communication with its customers, so that all parties can review the process data before the end of the aeration phase, thereby taking full advantage of the potential for timely product release.
For manufacturers, there may be a slight disadvantage to the implementation of parametric release, since EtO sterilizers may seek to recoup the costs of their new equipment by charging higher fees for each load released parametrically. On balance, however, the benefits of parametric release should justify these charges, which should contribute to the overall reduction of health-care costs.
A key element in the movement toward parametric release of EtO-sterilized products was put in place at AAMI's June 1995 meeting in Washington, DC, where three task groups were formed for the purpose of writing a technical information report (TIR) concerning various parts of ISO 11135. One of these groups is dedicated to writing a report on the engineering aspects of industrial EtO sterilization, which will include a section of complete guidance for anyone intent on initiating parametric release of EtO-sterilized devices. As currently conceived, this section is expected to discuss requirements for equipment, validation, routine controls and monitoring, release of product, and contract sterilization.
A first draft of the entire TIR--including the section on parametric release--was presented for committee review at another AAMI meeting last September. While there is still a great deal to be discussed--and the document is sure to undergo many more changes before final presentation to the full committee--it is not too early to discern some of the key issues related to parametric release now under discussion.
The September draft of the TIR stresses the importance of ensuring even distribution of moisture and sterilant, and emphasizes the need to monitor the actual functioning of recirculation systems during the preconditioning, sterilization, and aeration phases. In a related discussion, members of the task group questioned the recommendations of ISO 11135, Annex D (an informative appendix, and therefore not part of the standard's requirements), which suggests that product temperature across the load be permitted to vary only 3°C (5.4°F) above the minimum acceptable validated temperature. This means that if the sterilization cycle is validated to a run at a minimum of 120°F, the product temperature spread across the load should be controlled to a range of 120°125.4°F in order to satisfy the requirements for parametric release. Such a tight temperature range is realistic provided that the sterilization vessel is small, but members of the group noted that achieving such a tight product temperature spread in large industrial vessels would render parametric release prohibitively expensive or even impossible. As an initial solution, the task group proposed that its draft recommend a slightly higher acceptable temperature spread than that suggested by ISO 11135, Annex D. Whether this proposal will withstand further discussion remains to be seen.
Other issues of importance include product load configuration and the presence of "cold spots" within the load. Because a lower-than-validated product temperature would compromise the SAL delivered by the process, all members of the task group agreed that the minimum specified product temperature must be validated, maintained, and monitored at all times. But there was disagreement over the question of whether the product temperature range specified by previous standards (AAMI ST27 accepted a range of 18°F) needed to be narrowed for parametric release. Those favoring a less restrictive range reasoned that only the minimum temperature needed to be defined in order to ensure the product's SAL; the maximum temperature need only be governed by the thermal sensitivity of the product and packaging. Since higher temperatures accelerate and benefit the sterilization process, they should not be discouraged unless the product or packaging integrity are at risk. Although this question is yet to be resolved, one proposal was for the TIR to require only that the load be kept above the minimum validated temperature, without specifying a particular product temperature range.
Because product load configuration is an important element in the success of parametric release, the AAMI task group determined that manufacturers should be given more guidance on this subject than has previously been available. In the case of a load made up of only one type of product, it often happens that a certain location inside the vessel consistently yields lower temperatures. Such cold spots are typically found in pallets placed next to sterilizer doors, and may be due to recirculation factors affecting that particular area. In such cases, the group agreed that it is essential to identify such cold spots during the validation phase (using both empty-chamber qualifications and product/process validation cycles), and that additional probes be inserted in these locations during all routine production cycles.
In the case of loads made up of many different products (i.e., custom kits) the committee insisted that there be clear guidance regarding the management of cold spots. The validation should be engineered in such a way as to identify not only "fixed" cold spots (consistent with sterilizer location), but also "wandering" cold spots that originate from the presence of certain types of product within the load. Although the nature of the TIR's guidance is not yet determined, one proposal is to require that manufacturers conduct a comparative thermal study of all items composing the product family in order to determine which products show slow heat absorption or poor heat retention. Subsequent validation would then be engineered to include a worst-case load configuration of products deemed to be the most difficult to heat. Later, if a routine processing configuration contained a significant number of these worst-case products, additional temperature probes would be used to ensure compliance with the minimum validated temperature.
It is expected that the AAMI TIR will be published sometime in 1996, and it will be interesting to see how the document develops and what effect it will ultimately have on industry. Parametric release is not the only subject dealt with in the engineering TIR. Guidance will also be provided for all equipment necessary for proper EtO sterilization, for methods of calculating the relative humidity inside a vessel, and for dealing with issues of process equivalency among multiple vessels. There are many other ideas that may find their way into the document or into practice:
* Using a period of conventional product release with extensive temperature probing in order to accumulate thermal profiles of diverse product lot configurations.
* Employing various methods of gas analysis (flame ionization detector gas chromatography, Fourier-transform infrared spectroscopy, microwave molecular rotational spectrometry).
* Employing various methods of moisture analysis (electronic humidity sensors, thermal conductivity detector gas chromatography, Fourier-transform infrared spectroscopy, microwave molecular rotational spectrometry).
* Implementing a required frequency for headspace analyses (e.g., once at the beginning and once at the end of dwell, at five-minute intervals during the entire dwell).
* Requiring sterilizers to correlate calculated gas and moisture measurements with those verified through direct analysis.
* Establishing the need to monitor and control the quality of steam used during the conditioning phase.
The ultimate evolution in modern EtO sterilization of medical devices, however, is still some way off. This will come when routine parametric release is combined with in-chamber dynamic environmental conditioning, a process that bypasses the time-consuming phase of external preconditioning. When that practice becomes routine, products that are able to withstand a deep vacuum cycle will be transferred directly from the truck to the vessel, sterilized, aerated, and shipped to market--sometimes in less than 24 hours. The current developments in EtO sterilization are a small step forward along the path to such a future.
1. Guideline for Industrial Ethylene Oxide Sterilization of Medical Devices, ANSI/AAMI ST27, Arlington, VA, Association for the Advancement of Medical Instrumentation (AAMI), 1988.
2. Medical Devices--Validation and Routine Control of Ethylene Oxide Sterilization, ANSI/AAMI/ISO 11135, Arlington, VA, AAMI, 1994.
Paul J. Sordellini is a sterilization consultant with Quality Solutions, Inc. (Annandale, NJ).