Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI September 1997 Column

Contamination of imported and domestic cotton products prompts investigation into effectiveness of EtO sterilization.

During the summer of 1993, the Canadian Health Protectorate Branch found that some sterile laparotomy sponges from a U.S. manufacturer were contaminated with a fungal growth (or mold), later identified as Pyronema domesticum. The sponges, which had undergone EtO sterilization, were made from cotton that had been imported from China. Alerted to the situation, FDA and various manufacturers conducted tests of other lap sponges and China-origin cotton products, such as operating room towels, gauze, and surgical drapes, and found additional instances of Pyronema contamination. By early 1994, several voluntary product recalls had been initiated, and manufacturers of cotton products were investigating the organism and its resistance characteristics further.^1,2 Although Pyronema is not recognized as a human pathogen and no infections were reported, this episode raised some serious concerns about current sterilization practices and potentially resistant microorganisms. As a result, on April 22, 1994, FDA's Center for Devices and Radiological Health (CDRH) distributed a memorandum addressed to "All Device Manufacturers/Repackers Using Cotton." Using terminology that included the words should and must, this document made a number of recommendations concerning sterilization validation for cotton medical devices. Upon receiving the memorandum, members of the sterilization and packaging subcommittee of the Health Industry Manufacturers Association (HIMA) decided to respond formally. Accordingly, a Pyronema Working Group was formed that included the authors of this article, and a written response was submitted to FDA on June 14, 1994.

Despite undergoing sterilization, some cotton products have been contaminated by Pyronema domesticum.

The working group also developed a draft screening procedure for Pyronema.³ Their reason for doing so was twofold: to provide a relatively simple protocol that manufacturers could use to detect the presence of Pyronema on their products, and to provide a consistent method that could be used to collect data to answer questions about the prevalence of Pyronema in cotton. As stated in the "Scope" section of the draft, the procedure was not intended as a replacement for bioburden testing, nor was it intended to replace a manufacturer's standard process validation protocol. The draft screening procedure further specified that the technique should be used for detecting naturally occurring Pyronema on product rather than organisms resulting from an artificial source, such as an inoculum, which may not be representative of Pyronema's sterilization resistance in its natural state on products.

The draft screening procedure was sent to FDA for review and comment and then distributed in a letter format to HIMA members and nonmembers on February 1, 1995.⁴ Since that time, six manufacturers have submitted the results of their screening studies to HIMA for summary and analysis. A total of 45 studies were performed by these manufacturers using 16 products (differentiated based on product description, manufacturing plant, and product weight) and three sterilization modes: EtO, moist heat, and radiation. In addition, one company shared the results of its validation studies on product inoculated with Pyronema domesticum. The remainder of this article summarizes the study results and presents HIMA's recommendations regarding the sterilization of cotton medical products.

BIOBURDEN TESTING

Based on accepted practices,⁵ the screening procedure recommended that, whenever possible, the entire product should be used for bioburden testing rather than a sample item portion (SIP); that is, the SIP should equal 1.0. When testing the entire product is not practical, a study should be performed using various-sized sample portions to determine the minimum SIP needed to achieve bioburden recovery. All six manufacturers that submitted data followed the recommendation in the Pyronema screening procedure and tested whole products.

The bioburden recovery procedures used by some manufacturers were reported as validated, while others did not provide this information. The bioburden recovery validation distribution for the 16 products tested is shown in Table I. For those manufacturers that reported validating their procedures, bioburden recovery varied from 30 to 100%, with an average of 71.4%.

Table I. Bioburden recovery validation distribution for the 16 products tested.

The bioburden distribution on products used for sterilization studies, which is shown in Figure 1, was determined using the bioburden numbers as submitted. For those companies that reported data separately for bacteria and molds, the numbers were added for total bioburden. (Note: For those products in which a Pyronema recovery factor was reported, it is not known whether that recovery factor was used in the reported bioburden.) The data are shown in the figure in terms of bins, which represent various bioburden levels. Each bin is labeled in terms of its mean value (1 X 10^x), and the range on each bin is plus-or-minus half a log (for example, the bin labeled 10² ranges from 50 to <500, and the one labeled 10³ ranges from 500 to <5000).

Figure 1. Bioburden distribution data as submitted by the participating manufacturers.

The study results indicated that bioburden testing is of limited use in determining the presence of Pyronema. Unlike an immersion test performed following a partial sterilization cycle, which is capable of isolating Pyronema, there was no general consensus on isolation of Pyronema using bioburden testing.

ETO STERILIZATION TESTING

Screening Studies. The six manufacturers performed a total of 29 studies in which 1950 units of 13 product types were sterilized using a sublethal EtO process. According to the cycle descriptions provided, 100% EtO or 8.6:91.4% EtO: HCFC (hydrochlorofluorocarbon) mixtures were used in concentrations ranging from 535 to 800 mg/L. All exposure temperatures were between 125° and 135°F with exposure times of 30 to 45 minutes for 28 of the studies and a 9-hour exposure for one study. All testing was performed using soybean-casein digest broth (SCDB) as the culture medium with incubation temperatures at two different ranges: 20°–25°C (25 studies) and 28–­32°C (4 studies). Incubation times ranged from 14 to 30 days, but all reported data indicated that Pyronema domesticum survivors were detectable within 14 days. A summary of the test data is shown in Table II.

Table II. Summary of EtO sterilization screening study data.

The percentage of Pyronema positives was compared to the associated total bioburden count for each test group (in those cases where both pieces of information was provided); those data are shown in Figure 2. No correlation should be drawn between high total counts and high levels of Pyronema positives. The results might have been caused by the test units having come from several manufacturers. The source and processing of the raw materials as well as the procedures used to manufacture the products may have contributed to Pyronema remaining as part of the natural bioburden until sterilization processing. Results also suggest that more Pyronema survivors occur when sterilized products are incubated at temperatures between 20° and 25°C. It should be noted, however, that all 28°–32°C testing was performed by one manufacturer, which did not provide information on the rate of Pyronema detection. Overall, the reported data indicate that EtO may not be an effective mode of sterilization for product contaminated with resistant strains of Pyronema.

Figure 2. Results of EtO screening studies, showing Pyronema positives as a percentage of total counts.

Of the 29 EtO sterilization screening studies, 26 involved products made from cotton obtained from China and the remaining 3 were of products using cotton obtained from the United States. These studies are the first documented evidence of Pyronema in cotton from a country other than China. The organism in the contaminated domestic cotton cases was identified as Pyronema spp., not Pyronema domesticum.

Inoculated-Product Study. In the validation study submitted by one manufacturer, Bacillus subtilis biological indicators (BIs) and product inoculated with Pyronema domesticum (1100 ascospores and 10 sclerotia) were subjected to a 6-hour EtO sterilization cycle with a gas concentration of 450 mg/L. As in the screening studies using uninoculated product, Pyronema growth was detected within 7 days. All 10 of the product units tested produced Pyronema growth, while all of the BIs were negative for growth. These data reinforce the premise that Pyronema is more resistant to EtO than Bacillus subtilis BIs. However, caution should be used when attempting to relate the resistance of an organism as it naturally occurs to that of its lab-cultured counterpart.

MOIST-HEAT STERILIZATION TESTING

Screening Studies. Only two manufacturers provided screening results on sublethal moist-heat sterilization processing. One manufacturer tested a total of 1200 units (representing four types of laparotomy sponges) that had been exposed to multiple sterilization cycles with F_°s ranging from 21.2 to 37.0 minutes. The units were incubated in SCDB for 14 days at 20°–25°C. The other manufacturer tested a product that had a 58% Pyronema domesticum contamination rate in EtO screening studies. One hundred product units that had been exposed for 5 minutes at 250°F were incubated in SCDB at 20°–25°C for 30 days. No Pyronema growth was detected in either study.

Inoculated-Product Studies. Results also were submitted for two moist-heat studies in which products had been inoculated with Pyronema domesticum ascospores and sclerotia. Forty product units and standard moist-heat BIs (Bacillus stearothermophilus) were exposed for 10 minutes at 255°–260°F, and 72 product units were exposed for 5 minutes at 255°–260°F. Product units from the 10-minute cycle were split between SCDB and fluid thioglycollate medium and incubated at 20°–25°C and 30°–35°C, respectively. All product units from the 5-minute cycle were incubated in SCDB only. After 14 days of incubation, Pyronema domesticum was not detected in any of the product samples and all of the BIs were negative for growth.

RADIATION STERILIZATION TESTING

Screening Studies. Three manufacturers submitted results from radiation processing screening studies. Data from two of the manufacturers are summarized in Table III. In those two studies, 250 product units were exposed to radiation doses ranging from 7.6 to 12.8 kGy. All units were incubated in SCDB for 28 days, and no Pyronema growth was detected.

Table III. Summary of radiation sterilization screening study data from two manufacturers.

The third manufacturer performed studies in which two batches of 100 product units each were exposed to ever-increasing doses of radiation, followed by an EtO processing step to remove any organisms other than Pyronema that might have survived the radiation. The EtO processing was performed with 100% EtO at a concentration of 600 mg/L. The exposure temperature ranged from 120° to 130°F with an exposure time of 45 minutes. Processed product units were incubated in SCDB at 20°–25°C for 14 days. The results of these studies are shown in Table IV.

Table IV. Summary of data for screening studies that combined gamma radiation and EtO processing.

Inoculated-Product Studies. The results of a radiation sterilization study using inoculated product units were submitted by one manufacturer. The product was inoculated with Pyronema domesticum, but no data were included as to the number or form (sclerotia or ascospore or both) of inocula used. After exposure to a radiation dose of 19.2 to 27.8 kGy, 14 out of 60 units (23.3%) incubated in SCDB were positive for Pyronema. Growth was detected within 7 days. As mentioned previously, however, it is difficult to relate the resistance of lab-cultured inocula to that of naturally occurring contamination by the same organism.

RECOMMENDATIONS

In response to the issues raised in FDA's April 22, 1994, memorandum, HIMA has issued recommendations covering the areas of bioburden testing, sterilization cycle validation, and routine sterilization processing. The following recommendations apply to all cotton products.

Bioburden. All bioburden recovery methods should demonstrate the ability to recover a broad spectrum of microorganisms (bacteria, yeast, and molds) from the product. The specific bioburden recovery method chosen for a given product should be validated, including validation of the chosen SIP if whole-product testing cannot be performed.⁵ Bioburden should be assessed for products that are being used for sterilization cycle development studies. Periodically thereafter, bioburden should be assessed as part of the revalidation program.

Sterilization Cycle Development/ Validation. Products should be sterility tested as part of cycle development or validation. As with bioburden recovery methods, sterility test methods should demonstrate the ability to recover a broad spectrum of microorganisms from the product, and for testing using an SIP less than one, the chosen SIP must be validated.⁵ For EtO and moist-heat cycle development and validation, placement of the microbial challenge should be in the hardest-to-sterilize location within the chamber. The 14-day incubation period currently used for immersion/sterility testing of cotton products is sufficient to detect Pyronema after exposure to a partial sterilization process.

Manufacturers intending to use EtO sterilization should perform a screening study to determine the presence (or absence) of Pyronema domesticum in their cotton products. Sterilization of cotton products that have been shown in laboratory screening studies to be Pyronema free should be validated in accordance with ANSI/AAMI/ISO 11135.⁶ EtO sterilization may not be effective for cotton products containing Pyronema unless the cotton has been pretreated by a validated method effective against Pyronema domesticum. This pretreatment should be included as a part of the sterilization validation package (as part of the performance qualification). Sterilization of cotton products with radiation or moist heat should be validated in accordance with ANSI/AAMI/ISO 11137 and ANSI/AAMI/ISO 11134, respectively.^7,8

Routine Sterilization. As stated above, EtO sterilization may not be effective for cotton products containing Pyronema unless the cotton has been pretreated by a method effective against Pyronema domesticum. Once this requirement has been met, or if laboratory screening has shown the cotton product to be Pyronema free, the product can be effectively terminally sterilized using a validated EtO process.⁶ All cotton products can be effectively terminally sterilized using a validated radiation or moist-heat process.^7,8

CONCLUSION

The working group concluded that cotton products containing Pyronema domesticum can be successfully sterilized if the process is validated in accordance with current ANSI/AAMI/ISO standards.

Manufacturers interested in obtaining a copy of the HIMA screening procedure can visit the association's Web site at or fax a request to Stacey Robertson at HIMA at 202/783-8750.

REFERENCES

1. Bryans T, and Aaronson J, "Industry's Struggle with Pyronema," Med Dev Diag Indust, 16(10):102–109, 1994.

2. Freiherr G, "Mold Impairs Sterile Cotton Industry," Med Dev Diag Indust, 16(4):10, 16–20, 1994.

3. "Screening Procedure for Pyronema" (draft), Washington, DC, Health Industry Manufacturers Association (HIMA), 1994.

4. Letter containing "Screening Procedure for Pyronema," Washington, DC, HIMA, February 1, 1995.

5. Microbiological Methods for Gamma Sterilization of Medical Devices, AAMI TIR no. 8, Arlington, VA, Association for the Advancement of Medical Instrumentation (AAMI), 1991.

6. Medical Devices—Validation and Routine Control of Ethylene Oxide Sterilization, ANSI/AAMI/ISO 11135, Arlington, VA, AAMI, 1995.

7. Sterilization of Health Care Products—Requirements for Validation and Routine Control—Radiation Sterilization, ANSI/AAMI/ISO 11137, Arlington, VA, AAMI, 1994.

8. Sterilization of Health Care Products—Requirements for Validation and Routine Control—Industrial Moist Heat Sterilization, ANSI/AAMI/ISO 11134, Arlington, VA, AAMI, 1993.

Joyce Hansen is affiliated with Sherwood/Davis and Geck; Carol Lampe is affiliated with Baxter Healthcare Corp.; Trabue Bryans is affiliated with Axios, a ViroMed biosafety company; Gerry O'Dell is affiliated with Johnson & Johnson Medical, Inc.; Arnold Shechtman is affiliated with Medline Industries, Inc.; Thelma Wilcott is affiliated with Becton Dickinson and Co.; and Ann Baldwin is affiliated with HIMA.

Copyright ©1997 Medical Device & Diagnostic Industry