Originally published June 1996
The International Organization for Standardization (ISO) standard for ethylene oxide (EtO) sterilization residuals sets new limits using a different basis than that used by FDA in its 1978 proposal. Published late last year, the standard uses health-based risk assessment to establish limits, which are expressed as maximum allowable doses to the patient.1 The standard also includes analytical procedures for determining residue levels in devices that have been validated through interlaboratory evaluation.
The European Pharmacopeia provides requirements for devices involved in the administration of drugs, such as syringes and administration sets. The requirements are expressed in ppm, that is, parts of residue (µg) per million parts of device (g). In December 1993, the Committee for Proprietary Medicinal Products (CPMP), which reports to the Commission of the European Communities (CEC), set up a working party on the quality of medicinal products. The working party published a note for guidance that includes suggested restrictions on the use of EtO for the manufacture and sterilization of medicines and proposes limits on the amount of EtO and ethylene chlorohydrin (ECH) (or other halogenated ethylenehydrin) in raw materials, finished product, and containers.2 This guidance is based on the "as low as reasonably acceptable" principle for establishing the proposed limits. The CPMP guidance sets the limits for EtO at 1 µg/g in raw materials and finished product and 1 µg/ml (container volume) for containers, and the corresponding values for ECH at 50 µg/g and 50 µg/ml, respectively.
In 1978, in response to a number of regulatory actions directed at EtO, including an EPA proposal3 and actions by the Occupational Safety and Health Administration to lower the permissible exposure limit (PEL) for workers potentially exposed to EtO, FDA published a notice in the Federal Register.4 In this proposed rule, FDA reviewed the toxicity data available at that time for the effects of EtO, ECH, and ethylene glycol (EG), and proposed maximum residue limits and maximum daily levels of exposure for these residues in drug products, and maximum residue limits in medical devices. The proposed limits for medical devices were expressed in ppm, and devices were categorized in various ways that were not always unambiguous.
The introduction to the new ISO standard (AAMI/ANSI/ISO 10993-7) notes that manufacturers should consider the use of alternative materials and sterilization processes during product development and design to minimize exposure to residuals. Clause 4 defines specific requirements for EtO residues in medical devices. The standard provides information concerning the derivation of the specific requirements as well as important background information and guidance in informative annexes. It is essential that users of this standard read all the information provided and understand how it applies to specific medical devices.
Subclause 4.1 describes general requirements for EtO residues and notes that these requirements are in addition to the biological testing required in other parts of ISO 10993. Furthermore, for devices sterilized with EtO, the standard requires that manufacturers meet all applicable ISO 10993-1 requirements and account for the EtO residual level at the time of release for each medical device. The results of the biological assessment of the device may dictate more- stringent limits than those set out in subclause 4.3, which are designed to protect against systemic effects. Manufacturers need to consider the possibility of acute localized effects, particularly irritation, for all devices, especially small ones.
The biological evaluation of medical devices is a risk-assessment process to evaluate safety, part of the quality system within which medical devices are manufactured. Manufacturers must document the output of this evaluation as an element of their due diligence decision-making process.
Subclause 4.2 categorizes devices into limited exposure, prolonged exposure, and permanent contact. This is based on exposure durations defined in subclause 5.2 of ISO 10993-1--1992. If a device falls into more than one exposure duration category, manufacturers must apply the more rigorous requirements. With multiple exposures, manufacturers can determine the appropriate category based on the potential cumulative effect and the exposure time.
Subclause 4.3 delineates the maximum allowable doses of EtO and ECH delivered to patients for each exposure category. During the development of the standard, risk assessment for EG indicated that when manufacturers control EtO residues as required by this standard, it is unlikely that biologically significant EG residues will be present.
The limits for EtO and ECH are defined for permanent-contact, prolonged-exposure, and limited-exposure devices. For permanent-contact devices, a patient's average daily dose of EtO must not exceed 0.1 mg per day. In addition, a patient's maximum EtO dose must not exceed 20 mg in the first 24 hours, 60 mg in the first 30 days, and 2.5 g in a lifetime. A patient's average daily dose of ECH must not exceed 2 mg per day. In addition, a patient's maximum ECH dose must not exceed 12 mg in the first 24 hours, 60 mg in the first 30 days, and 50 g in a lifetime.
For prolonged-exposure devices, a patient's average daily dose of EtO must not exceed 2 mg per day. In addition, a patient's maximum EtO dose must not exceed 20 mg in the first 24 hours and 60 mg in the first 30 days. A patient's average daily dose of ECH must not exceed 2 mg per day, and the maximum ECH dose must not exceed 12 mg in the first 24 hours or 60 mg in the first 30 days.
For limited-exposure devices, a patient's average daily dose of EtO must not exceed 20 mg per day, and the average daily dose of ECH must not exceed 12 mg per day. Manufacturers need to be aware that the simultaneous use of more than one device and the use of devices for treating neonates may result in additional exposure. This issue is described in annex E of the standard.
The standard expresses the limits for permanent-contact and limited-exposure medical devices as maximum average daily doses. These limits carry additional constraints for the first 24 hours of the exposure period and, in the case of permanent-contact devices, additional constraints for the first 30 days. These constraints limit the amount of EtO and ECH that can be delivered to patients during these early periods.
During development of this standard, the committee recognized some special situations in which the limits given in subclause 4.3 would not be practical. Special situations could include limitations of the devices themselves or cases in which human data required more stringent limits. Such situations include:
* For multidevice systems, the limits apply to each individual device.
* For intraocular lenses, EtO residue shall not exceed 0.5 µg per lens per day, nor 1.25 µg per lens.
* For blood oxygenators and blood separators, a patient's average daily dose of EtO shall not exceed 60 mg.
* For extracorporeal blood purification set-ups, the EtO and ECH limits specified for prolonged- and limited-duration medical devices apply, but the allowable dose for a lifetime may be exceeded.
DETERMINING RESIDUE LEVELS
With any analytical method, it is crucial for manufacturers to select a representative sample to measure. This is, of course, true for determining EtO residues in medical devices to demonstrate compliance with the standard. In evaluating how to obtain a representative sample, manufacturers must consider a number of factors listed in annex C of the standard.
Sample Handling. If an analyst takes samples to generate a decline curve, it may be necessary to replace or backfill the pallet, carton, or shipper after removing a sample from the product load. Manufacturers must take adequate precautions to protect the analyst, and extract or ship product as soon as possible after removing the sample from the product load.
If a manufacturer ships samples to a remote laboratory for analysis, it is the manufacturer's responsibility to transfer the sample to an appropriate extraction container and commence extraction, or to seal the sample in a shipping container with sufficient dry ice to prevent dissipation of residual EtO during shipment.
To ensure that no other sample matrix components have the same gas chromatographic (GC) retention time as the residues being determined, the analyst should use the identical procedure to extract a nonsterilized sample. This sample will act as a control, or blank. Modifying the chromatographic conditions may be necessary to separate any interfering peaks from the analyte peak; if this cannot be accomplished, the manufacturer may need to substitute an alternative analytical procedure.
Extraction Procedures. The standard describes two residue extraction methods: simulated-use and exhaustive extraction. Because simulated-use extraction is the only method that produces results directly comparable to the specified limits, this is the reference method. All of the methods described in annex B of the standard have been validated through interlaboratory studies.
Manufacturers that use aqueous simulated-use extraction must ensure that conditions provide the greatest challenge to the device's intended use. For example, analysts can extract many blood-contacting or parenteral devices by filling or flushing the blood or fluid path for at least the maximum time for single use or for a time that ensures total extraction at temperatures that provide the greatest realistic simulated challenge. Analysts can use extraction rates to calculate effects of longer or repeat exposure.
Simulated-use extraction may not extract the entire amount of residue in the device. Manufacturers must be aware that using water may convert EtO to EG or, where chloride ions are present, EtO to ECH. Analysts must evaluate this possibility and take steps to control any adverse effects. They must validate any simulated-use extraction procedure to demonstrate the actual exposure level to patients.
The standard allows manufacturers to use exhaustive extraction to recover the entire residue content, provided they address the shorter-term constraints. These extraction procedures include thermal extraction followed by headspace-gas analysis and solvent extraction procedures, with either headspace-gas analysis of the solvent extract, chromatography of the solvent extract, or preparation of the bromohydrin derivative of EtO, which is determined using a more sensitive GC detector.
The standard specifies GC methods, validated through interlaboratory study, for the quantitation of residues (see Table I). The standard discusses the evaluation of gas chromatograms in normative annex A and specifies requirements for resolution, tailing, relative standard deviation of the standard curve, and the chromatographic baseline. The generation of reliable and reproducible GC data depends on adherence to these minimum system suitability requirements.
Because toxicity is a function of dose, the ISO standard specifies the maximum allowable dose as the criterion for limiting patient exposure to EtO residues.
According to the ISO standard, a manufacturer can release a product for market when residues are less than those specified in the standard, "provided that all other applicable biological testing requirements of ISO 10993-1 have been met when the device contains EtO residues at the level specified by the manufacturer at the time of release." If a manufacturer has sufficient data on residue diffusion kinetics, it may be possible to group devices for quality assurance testing based on similarity of materials, manufacturing processes, and end uses.
Analysts can use dissipation curves to estimate the poststerilization time for products or families of similar products to reach residue limits. If aeration temperatures differ, data can be pooled from sterilization loads taken from aeration or quarantine storage at different times of the year. Manufacturers must also consider the presence of adjacent EtO-sterilized medical devices when obtaining experimental data for dissipation curves.
Regression analysis of pooled data from sufficient time points for at least three lots of the same product will establish the nature of the dissipation curve, enabling the device being tested to be released at the calculated upper 95% prediction limit for it.
Manufacturers must obtain all data for release of medical devices in compliance with this part of ISO 10993 from experiments and data analyses according to valid standard operating procedures. When the sterilization process parameters change, manufacturers should reevaluate product residues to determine their next course of action.
FACTORS AFFECTING RESIDUES
To properly analyze residues in EtO-exposed devices, analysts must recognize parameters that affect residue content. Sufficient experimental evidence on residue diffusion kinetics--that is, the rate of EtO gas diffusion from the various device components--may make it possible for a manufacturer to address a family of devices through the analysis of a worst-case representative, precluding the need to analyze each product in the line. Although devices may be related based on size and use, material composition, packaging, EtO exposure, water content, or exposure to environmental conditions, variance in these factors could lead to different residue levels.
Material Composition. Materials vary considerably in their ability to absorb, retain, and release EtO. When conversion of EtO to ECH is possible, two similar devices made of different materials are likely to have very different residue profiles. The concentration of ECH varies greatly in materials that contain a source of free chloride ions. A single device composed of two dissimilar materials may require a representative sample of both materials to ensure accurate analysis. A manufacturer may need to consider composition and size to simulate normal product use.
Packaging. Packaging materials vary widely in their abilities to allow penetration and dissipation of both EtO gas and other possible residues, which may in turn affect ECH residue levels. Packing density and the density of the shipping container are other sources of variability.
EtO Exposure. Process conditions under which a device is exposed to EtO will affect the residue levels. These conditions include gas concentration, exposure time, temperature, type of cycle (pure EtO or EtO mixtures), humidity (including the quality of the water source), reevacuations and air washes, product and load density, or the configuration of the product load in the sterilizer.
Residual EtO in devices is also a function of aeration conditions, including temperature, load density and configuration, airflow, loading pattern, surface area of products being aerated, and aeration time. Some materials demonstrate aeration rates that can roughly double (aeration time reduced by one-half) for each 10°C increase in aeration temperature. Factors such as humidity, temperature, and airflow may influence ECH formation depending on EtO content in the product after removal from the sterilizer.
Environmental Conditions. Analysts need to be aware of seasonal variations in aeration rates when samples are stored under laboratory conditions that differ from ambient warehouse conditions. Under certain circumstances, analysts should hold samples prior to analysis under conditions that approximate the lowest temperature at which the product is likely to be stored.
Operators should exercise caution when removing product samples from the sterilization load soon after the sterilization process is completed, or when product samples or extracts thereof are shipped to an analysis site. In such cases, attempting to correlate the residue amounts on samples and on the rest of the load may result in errors; operators should conduct experiments to establish the relationships between these sets of conditions.
REVISION OF ISO 10993-7
The International Agency for Research on Cancer working group on the evaluation of carcinogenic risks to humans, which met in Lyon, France, February 1525, 1994, reclassified EtO to Group 1, defined as carcinogenic to humans. This reclassification caused concern to those developing the ISO standard at that time. The participating, or voting, members of ISO/TC 194, ISO/TC 198, and CEN/TC 206 (as well as FDA in the United States) voted to accept the standard because the absence of a standard was unacceptable, but all parties agreed to start immediately revising it.
The working group proposed a timetable for revising the standard to address concerns of a number of the ISO and European Committee for Standardization (CEN) member bodies, as well as FDA. The new standard is due in June 1998. As background for this activity, the working group proposed a new work item to develop a guidance for the procedure to develop residue limits for any residue using health-based risk assessment. The technical committee accepted this proposal. The working group also proposed a timetable for revising the requirements and methods in clause 4 and annexes B and E of the standard. Working group members have agreed that ISO CD 14538, the method for establishing allowable limits for residues in medical devices using health-based risk assessment, can be used as the rationale for the EtO-residue standard. This EtO work will proceed once an ISO CD 14538 method guidance is developed.
GUIDANCE FOR ISO 10993-7
The U.S. delegation to the working group proposed the development of guidance for this standard in 1993. At that time, the proposal was not seconded, so no discussion took place. In the development of the U.S. position and ballot, FDA proposed that guidance was needed and made such a guidance a condition for its acceptance of the U.S. position. The U.S. subtechnical advisory group formed a working group composed of industry and FDA members to develop such guidance.
Members of the ISO working group did not agree that guidance was needed because they believed that the standard contained sufficient information to enable compliance to be easily achieved. Because people seem to find aspects of this standard confusing, a guidance document would probably be helpful. It is important that people continue to provide input that will help explain the standard, its requirements, and how manufacturers can comply.
1. "Biological Evaluation of Medical Devices-- Part 7: Ethylene Oxide Sterilization Residuals," ANSI/AAMI/ISO 10993-7, Arlington, VA, Association for the Advancement of Medical Instrumentation, 1995.
2. "Limitations to the Use of Ethylene Oxide in the Manufacture of Medicinal Products," Brussels, Belgium, Commission of the European Communities, December 1993.
3. "Notice of Rebuttable Presumption against Registration and Continued Registration of Pesticide Products Containing Ethylene Oxide," Federal Register, 43 FR: 38013815.
4. "Ethylene Oxide, Ethylene Chlorohydrin, and Ethylene Glycol--Proposed Maximum Residue Limits and Maximum Levels of Exposure," Federal Register, 43 FR: 2747427483.
Barry Page is co-chair of the AAMI ethylene oxide residuals working group and convenor of ISO technical committee 194 working group 11, "Biological Evaluation of Medical Devices: Sterilization Process Residuals."