Abstract
Current practice in National Health Service (NHS) hospitals employs 70% Industrial Methylated Spirit spray for surface disinfection of components required in Grade A pharmaceutical environments. This study seeks to investigate other agents and procedures that may provide more effective sanitisation. Several methods are available to test the efficacy of disinfectants against vegetative organisms. However, no methods currently available test the efficacy of disinfectants against spores on the hard surfaces encountered in the pharmacy aseptic processing environment. Therefore, a method has been developed to test the efficacy of disinfectants against spores, modified from British Standard 13697 and Association of Analytical Chemists standards.
The testing procedure was used to evaluate alternative biocides and disinfection methods for transferring components into hospital pharmacy cleanrooms, and to determine which combinations of biocide and application method have the greatest efficacy against spores of Bacillus subtilis subspecies subtilis 168, Bacillus subtilis American Type Culture Collection (ATCC) 6633, and Bacillus pumilis ATCC 27142. Stainless steel carrier test plates were used to represent the hard surfaces in hospital pharmacy cleanrooms. Plates were inoculated with 107–108 colony-forming units per milliliter (CFU/mL) and treated with the various biocide formulations, using different disinfection methods. Sporicidal activity was calculated as log reduction in CFU.
Of the biocides tested, 6% hydrogen peroxide and a quaternary ammonium compound/chlorine dioxide combination were most effective compared to a Quat/biguanide, amphoteric surfactant, 70% v/v ethanol in deionised water and isopropyl alcohol in water for injection. Of the different application methods tested, spraying followed by wiping was the most effective, followed closely by wiping alone. Spraying alone was least effective.
Introduction
Aseptic processes are vital in the hospital pharmacy environment, where sterile products are compounded to meet individual patients' specific needs. Aseptic dispensing is performed in a Grade A environment, such as a laminar flow cabinet or an isolator where pharmaceutically active but heat-labile products are aseptically prepared. Ineffective transfer techniques can therefore compromise the sterility of formulations produced under these conditions and can ultimately lead to nosocomial infection. One of the major sources of contamination during hospital aseptic processing is surface bioburden (1). A recent study by Stubbs (2) established that 60% of items that are frequently passed into pharmaceutical isolators are contaminated with bacteria. Disinfection is therefore required at transfer stages of aseptic processing to eliminate such contamination risks.
A number of different chemical disinfectants are currently employed in hospital aseptic processing. Disinfectants used for cleaning hard surfaces in hospital pharmacy and industrial cleanrooms are generally alcohol-based agents (1). Current practice in the majority of National Health Service (NHS) hospitals is to use 70% industrial methylated spirit (IMS), applied as a spray or wipe (3–5). Although alcohols kill a broad range of vegetative organisms, they have no activity against spores (6–8). Spores produced by environmental bacteria are a major concern in hospital pharmacy aseptic areas (1–3, 9–13). Such spores can withstand extremely harsh, unfavourable environments and are among the most resistant of all living cells to biocides (7, 14). Spore structure plays an important role, as the spore coat, cortex, and inner membrane form some of the spore components contributing to biocide resistance (6, 7, 15–20). Only a small number of biocides are active against both vegetative organisms and spores. These include glutaraldehyde, peroxygens, and halogens (6–8, 15, 21). However, these agents are damaging to both humans and equipment, thus there is currently no disinfectant which meets all the requirements of an effective sporicidal biocide.
A number of different chemical disinfection methods are employed in hospital aseptic disinfection of pharmaceutical components. The main disinfection methods used are spraying with a biocide, wiping with a biocide-impregnated wipe, and combinations of either spraying followed by wiping, or wiping followed by spraying (1, 5). One of the most commonly used hospital aseptic disinfection transfer systems consists of a combination of spraying with sterile 70% alcohol and wiping with sterile disinfectant-impregnated wipes, for example, isopropyl alcohol (IPA) wipes (3, 5).
No standard methodology currently exists for the comparison of disinfectants and disinfection procedures against spores present on hard surfaces. Association of Analytical Chemists (AOAC) and British Standard European Norm (BSEN) standards are only available for the activity of disinfectants applied against spores in suspension. Suspension testing does not represent the manner in which disinfectants are applied in the cleanroom environment. Due to the absence of a hard-surface test for testing disinfectants with sporicidal activity, a combination of BSEN 13697:2001 and AOAC methods was used to determine the efficacy of different disinfectants and disinfection methods against spores. This new method was developed to represent the working practice of a hospital aseptic dispensing unit. It takes into consideration the number and composition of the components and the cleanroom and isolator surfaces and composition.
This study was designed to produce a method which could be used to evaluate the sporicidal activities of currently available cleanroom biocides and their different methods of disinfectant application. A variety of disinfectants were applied in various formats to artificially contaminated hard surfaces to determine the effectiveness of each biocide and transfer disinfection method.
Materials and Methods
Materials
Non-pathogenic strains of Bacillus subtilis subspecies subtilis 168 spores (standard laboratory strain), Bacillus subtilis (ATCC 6633) spores, and Bacillus pumilis ATCC 27142 spores were obtained from the Culture Collection in the Microbiology Department at Aston University, Birmingham, UK.
The standard laboratory strain and the ATCC strain of B. subtilis were employed to address whether there was any intraspecies variation in susceptibility to the biocides. All the disinfectants tested in this study (Table I), apart from the non-fleecing cotton face pads (Boots, Nottingham, UK), were manufactured by Shield Medicare (Farnham, Surrey) according to Good Manufacturing Practice (GMP) and were applied at their in-use concentration.
Compositions of the Different Biocides Used
Tryptone soya agar (TSA) and tryptone soya broth (TSB) were from Oxoid (Hampshire, UK). Sodium thiosulphate, L-histidine, asparagine, and fructose were from Sigma (Poole, UK). Glucose was from Fisher (Loughborough, UK). Modified Letheen broth was from Difco, Becton Dickinson (Oxford, UK), and saponin was from Fluka BioChemika (Steinheim, Germany).
Methods
Preparation of Bacillus Spores:
Organisms were grown aerobically in 200 mL of TSB on an orbital shaker at 37 °C for 12 days, until a high percentage of spores were produced (determined by phase contrast microscopy). The spore suspensions were heated to 60 °C for 10 min to remove any vegetative cells and centrifuged at 10,000 g for 30 min at 5 °C. The spore pellets were washed in 0.9% saline before resuspending in 100 mL of sterile distilled water (SDW). The final sporulated suspension was heated to 60 °C for 30 min and kept at 2–8 °C until use. As the purpose of this study was to test the sensitivity of Bacillus spores to the different biocides, it was vital that all vegatative cells were removed from the inoculum. To ensure that this was the case, two heating stages were used to kill off any heat-resistant vegetative organisms surviving the first heating stage and to ensure that any spores which may have germinated in the SDW were removed. A viable spore count was performed before use and spore concentrations in the range of 107–108 colony-forming units per millititer (CFU/mL) were used in the tests.
Biocide Neutraliser Validation:
Modified Letheen neutralising broth (4.4 g/100 mL) containing sodium thiosulphate (0.5 g/100 mL), L-histidine monohydrochloride (0.1 g/100 mL) and saponin (3 g/100 mL) was used as a general neutraliser for each of the biocides.
One milliliter of biocide was added to 9 mL of Letheen neutralizing broth and incubated for 10 min at room temperature. Biocide plus neutraliser, neutraliser alone and SDW (viability control) were inoculated with Bacillus spores and incubated for 10 min at room temperature. Recovery of the organisms was determined following overnight incubation on TSA at 37°C, with the percentage of viable cells calculated according to Sutton et al. (22):
Treatment with Disinfectants:
Stainless steel plates (8cm × 8cm) were used as carriers, to represent the components that are regularly used in hospital pharmacy aseptic dispensing. Prior to disinfection the carriers were washed and rinsed with distilled water, before sterilising by autoclaving at 121 °C for 15 min.
Spraying Method:
Stainless steel plates were wrapped in foil, warmed to 37 °C for 30 min then inoculated with 100 μl of stock spore suspension which was spread evenly over the surface using a hockey stick spreader. After 15-min incubation at 37 °C, each inoculated carrier was laid flat and dosed with one spray of the test disinfectant from a distance of 10cm at room temperature. The stainless steel carrier was allowed to dry at room temperature for a contact time of 2 min before aseptic transfer into a sterile stomacher bag (Seward, Leicestershire). Ten milliliters of sterile Letheen broth-based neutralising solution was added and the bag sealed and agitated in a stomacher machine. After agitation for 2 min, 0.1 mL of the neutraliser recovered spores were removed, added to 0.9 mL of asparagine (0.33 mg/mL), glucose (1 mg/mL), fructose (1 mg/mL) and potassium (3.3 mg/mL) (AGFK) germinant (18, 19) and incubated at 37 °C for 45 min. Following this incubation period, 0.1 mL of the recovered AGFK/neutraliser treated spores were added to 0.9 mL of sterile Letheen neutraliser and a dilution series in neutraliser carried out. 0.1 mL of each dilution was plated onto TSA incubated overnight and CFU enumerated. Log reductions of spores were calculated as follows:
where controls surfaces are inoculated but not disinfected.
Negative controls were stainless steel plates without inoculum or disinfectant.
During the spraying process, inoculated carriers were placed on paper towels which were then processed in the same way to determine any displacement effects of spraying.
Wiping Method:
Surfaces inoculated as spraying, however the inoculated carriers were wiped using a pre-impregnated disinfectant wipe. A new area of the wipe was used each time to prevent any possible cross-contamination. A 25% overlap was used to make sure that the whole of the stainless steel surface came into contact with the disinfectant. Following a 2-min contact time with the disinfectant any viable spores remaining were neutralised, stomached, added to AGFK germinant and recovered as described above. After wiping, the wipe itself underwent neutralisation, stomaching and recovery to determine, if any, the number of spores being released from the carrier on wiping.
Spraying and Wiping Method:
Surface inoculated as spraying, then carriers were sprayed for a contact time of 2 min, followed by wiping for a further contact time of 2 min. Following this, neutralisation, germination (AGFK) and recovery was carried out.
Suspension Testing Method:
Suspension testing of the disinfectants was carried out in parallel with the hard surface testing. 100 μl of the aqueous spore suspension was mixed with 0.9 mL of each disinfectant for 2 min at room temperature, then 100 μL of the disinfectant/spore suspension was added to 0.9 mL of Letheen neutraliser and serial dilutions in neutraliser carried out. Each dilution was then plated in triplicate onto TSA and incubated at 37°C overnight. CFU were counted and the log reductions in spore levels determined.
Statistical Analysis
A non-parametric statistical analysis was carried out on all data using an ordinary ANOVA (Kruskal-Wallis). Post-hoc testing using Dunn's Multiple Comparisons was used to compare methods of disinfection, as well as the disinfectants themselves. Data calculated as percentages was corrected using arcsin and then statistical analysis carried out. Data are expressed as mean ± standard error of the mean SEM. Differences were considered to be statistically significant when P < 0.05.
Results
Neutraliser Efficacy and Toxicity
The results of the neutraliser evaluation procedure in Table II indicate that the neutraliser had a small effect on the number of viable cells with up to 5% of the spores being killed in the presence of the biocide and neutraliser solution. The sensitivity observed is not entirely due to the biocide, because the neutraliser alone resulted in a 1–3% reduction in spore levels.
Percentage Recovery of Viable Spores after Neutralisation with Modified Letheen Broth-based Neutraliser. Percent recovery is calculated based on the number of spores inoculated, which were: B. subtilis ATCC 6633: 3.3 × 108 CFU/mL, B. subtilis 168: 5.4 × 108 CFU/mL, B. pumilis ATCC 27142: 6.5 × 107 CFU/mL. Data are represented as mean ± SEM, n = 3. P-values indicate that there were no significant differences between the percentage recovery with neutraliser only compared to any of the biocides tested.
The variation among the biocide/neutraliser and neutraliser alone is not significantly different to that of the SDW viability control. Therefore although the neutraliser used was found to be slightly detrimental to spore viability, it was established that the extent to which the neutraliser was active against the spores (all organisms) was not significant (P > 0.05) when considering biocide efficacy for this study. Therefore a validated neutraliser formulation was used to ensure that biocide activity was arrested after the required contact time, without the neutraliser being toxic to the spores. Incorporation of the hydrogen peroxide neutraliser catalase (1000 units/mL) to the neutraliser formulation did not have a significant effect on both neutraliser efficacy and toxicity (P > 0.05), and was therefore omitted from the neutraliser (data not shown).
Efficacy of Disinfectants against B. subtilis subsp. subtilis 168 Spores
Figure 1 shows that the suspension testing generally achieved higher log reductions—ranging from 0.21 to 2.07 against B. subtilis 168 spores—than did the hard surface testing. This was followed by the spraying and wiping combination, which achieved a highest log reduction of 1.72 with the Quat/chlorine dioxide. The spraying alone method achieved the lowest log reductions in spore levels. All disinfection methods tested differed significantly from each other, with the exception of the difference between spraying alone and suspension testing being non-significant (P > 0.05) against the B. subtilis 168 spores. Suspension testing in SDW did not result in any log reductions for any of the organisms (data not shown).
Comparisons between log reductions of B. subtilis subsp. subtilis 168 spores following 1) Spraying (black bars), 2) Wiping (grey bars), 3) Spraying and Wiping (open bars), and 4) Suspension (hatched bars) testing using six disinfectants and SDW. Data are not presented for suspension testing in SDW since this did result in any log reduction. Data are represented as mean ± SEM, n = 9. Overall, all disinfection methods differed significantly from each other (P < 0.05). However, spraying was not significantly different from suspension testing (P > 0.05).
Efficacy against B. subtilis ATCC 6633 Spores
Figure 2 shows that higher log decimal reductions in the range of 0.98 to 1.17 were achieved against the B. subtilis ATCC 6633 spores when a spraying followed by wiping method was used. Suspension testing achieved lower log reductions in comparison to the hard-surface test results. Differences between all disinfection methods were significant (P < 0.05), with the exception of the difference between wiping alone and spraying and wiping against the B. subtilis ATCC 6633 spores.
Comparisons between log reductions of B. subtilis ATCC 6633 spores following 1) Spraying (black bars), 2) Wiping (grey bars), 3) Spraying and Wiping (open bars), and 4) Suspension (hatched bars) testing using six disinfectants and SDW. Data are not presented for suspension testing in SDW since this did result in any log reduction Data are represented as mean ± SEM, n = 9. Overall all differences observed between the disinfection methods were found to be significant (P < 0.05). However, even though spraying and wiping showed an increased level of efficacy in comparison to wiping alone, the difference between wiping alone and spraying & wiping was found to be non-significant (P > 0.05).
Efficacy against B. pumilis ATCC 27142 Spores
For the B. pumilis ATCC 27142 spores (Figure 3), each biocide achieved higher log decimal reductions ranging from 0.92 to 1.61 when spraying followed by wiping was used as a method of disinfection. Differences between all disinfection methods were significant (P < 0.05), with the exception of the difference between wiping alone and spraying and wiping against the B. pumilis ATCC 27142 spores.
Comparisons between log reductions of B. pumilis ATCC 27142 spores following 1) Spraying (black bars), 2) Wiping (grey bars), 3) Spraying and Wiping (open bars), and 4) Suspension (hatched bars) testing using six disinfectants and SDW. Data are not presented for suspension testing in SDW since this did result in any log reduction. Data are represented as mean ± SEM, n = 9. Overall, all differences observed between the disinfection methods were found to be significant (P < 0.05). Once again, even though spraying and wiping showed an increased level of efficacy in comparison to wiping alone, the difference between wiping and the combination of spraying and wiping was found to be non-significant (P > 0.05).
Figures 1–3 show that all three methods of disinfection gave rise to different levels of log reductions in spores. Taking into account all of the biocides and all the test organisms, it was found that the difference in log reductions observed between wiping and the combination of spraying and wiping against the three Bacillus spores was non-significant (P > 0.05). In comparison, the differences in log reductions in spore levels obtained on suspension testing in relation to spraying with the biocides was established as being significant (P < 0.05).
Efficacy of Disinfection Methods
Figure 4 summarises the efficacies of the different disinfection methods against the three spore suspensions. Spraying with a disinfectant was the least effective of the disinfection methods, with an average log reduction of at least 0.45 logs. Slightly better than the spraying was the suspension testing, which achieved at least an average 0.49 log reduction against the Bacillus spores. Spraying and wiping produced the best results against the Bacillus spores, achieving at least an average 1.06 log reduction. This was closely followed by wiping, which achieved at least a 0.83 log reduction. Only the spraying and wiping combination method achieved an average of at least a 1 log reduction against all three of the Bacillus sp. spores tested. The spraying alone technique did not achieve at least a 1 log reduction against any of the test organisms investigated. Log reductions obtained by all four disinfection methods against all the Bacillus spores were found to be significantly different from each other (P < 0.05). This was with the exception of the spraying alone and suspension testing methods (P > 0.05). The most effective method of disinfection was spraying and wiping, followed by wiping alone.
Average log reductions achieved for each disinfection method employed against B. subtilis 168 (black bars), B. subtilis ATCC 6633 (grey bars), or B. pumilis ATCC 27142 (open bars) spores. Data are represented as mean ± SEM for the log reduction of each organism with every biocide tested (n = 9 replicates for each organism and biocide combination). *Denotes significant (P < 0.05) differences between suspension testing and spraying, wiping or spraying, and wiping for the Bacillus spore type indicated.
Efficacy of Disinfection Methods—SDW against All Bacillus Spores
Figure 5 shows that spraying with SDW was found to be the least effective of the disinfection procedures, with an average log reduction of at least 0.25 logs. Of the methods investigated using SDW, spraying and wiping produced the best results against the Bacillus spores, achieving at least an average 0.92 log reduction. This was closely followed by wiping alone which achieved at least a 0.87 log reduction. Of the disinfection methods investigated, both wiping alone and spraying and wiping achieved an average of at least a 1 log reduction against some of the three Bacillus sp. spores tested. The log reductions obtained by all three disinfection methods using SDW against all the Bacillus spores were found to be significantly different from each other (P < 0.05). This was with the exception of the wiping alone and the spraying and wiping test methods (P > 0.05).
Average log reduction achieved using SDW on spraying, wiping, spraying and wiping, and suspension testing disinfection methods against B. subtilis 168 (black bars), B. subtilis ATCC 6633 (grey bars), and B. pumilis ATCC 27142 (open bars) spores. Data are represented as mean ± SEM, n = 9. *Denotes significant differences between spraying and wiping alone or spraying and wiping (P < 0.05) for the Bacillus spore type indicated.
Spore Release from Hard-Surface Carriers
Table III shows that spraying washed spores off the stainless steel-inoculated carriers onto the underlying paper towel. The recovery of spores from the paper towel indicates that the observed loss of spores from the surface was due to physical removal after spraying rather than sporicidal effects of the biocides.
Mean Percentage of B. subtilis ATCC 6633 Spores Recovered from Paper Towel Underlying the Steel Carrier Following Spraying. 100% recovery would be the entire inoculum of 7.8 × 107 CFU/mL. Data are represented as mean ± SEM, n = 2 with three replicates.
The amphoteric surfactant and the SDW sprays removed the most spores (26.1 ± 0.22% and 20.8 ± 0.01%, respectively). Spraying with the Quat/chlorine dioxide resulted in at least 10% of the B. subtilis ATCC 6633 spores being removed. The Quat/biguanide, Quat/chlorine dioxide, 70% v/v IMS, and 70% IPA all achieved similar percentages of spores being washed off, ranging from 1.3% to 2.5%.
It was found that P > 0.05 for all the biocides tested. So the variation among the ability of sprays to wash off the spores from the carrier to the underlying surface did not differ significantly between each of the disinfectant sprays tested.
Spore Transfer to Wipes
The results in Table IV show that an average of at least 14.8% of B. subtilis subsp. subtilis 168 spores were wiped off from the inoculated stainless steel carriers onto the disinfectant-impregnated wipes. Of the B. subtilis ATCC 6633 spores inoculated onto the stainless steel surfaces an average of at least 19.5% were recovered from the disinfectant wipes. The B. pumilis ATCC 27142 spores were found to be transferred in large numbers from the inoculated stainless steel surface onto the disinfectant wipes, with at least 41.3% of spores recovered from wipes. Overall, the wipe with 70% IPA with water for injection (WFI) was found to harbour the largest number of transferred spores, with an average of at least 50% of spores being recovered from the wipes.
Mean Percentage of B. subtilis 168, B. subtilis ATCC 6633, and B. pumilis ATCC 27142 Spores Recovered from Disinfectant Wipes, Following Wiping. Data are represented as mean ± SEM, n = 2 for B. subtilis spores and n = 3 for B. pumilis ATCC 27142 spores. Percent recovery is calculated based on the number of spores inoculated, which were B. subtilis ATCC 6633: 7.7 × 107 CFU/mL, B. subtilis 168: 7.6 × 107 CFU/mL, and B. pumilis ATCC 27142: 6.3 × 107 CFU/mL.
It was found that P > 0.05 for all the disinfectants tested. So the variation among the wipes in relation to the transfer of spores from the carrier surface to the wipe itself was not significantly different between each of the disinfectant wipes.
Evaluation of Biocide Efficacy
Figure 6 shows a comparison of the activity of the different biocides. This approach was taken to try and define the most effective biocide for spore removal, independent of the method of application and the species of contaminating spore. For every application method, the average log reduction of the six biocides tested was calculated (Figure 6).
Average log reduction achieved by each disinfectant biocide for the three spore types and four disinfection methods employed. Data are represented as mean ± SEM, n = 9. *P < 0.05 differences in log reductions versus 6% hydrogen peroxide.
Some level of log reduction was observed on the application of all disinfectants on suspension testing, spraying, wiping, and on a combination of spraying and wiping (Figures 1–3). The Quat/chlorine dioxide and the 6% hydrogen peroxide showed the highest average log reductions (1.13 and 1.22, respectively). The Quat/biguanide, 70% v/v IPA in WFI and the SDW achieved similar average log reductions in spores. The amphoteric surfactant and the 70% v/v ethanol also achieved similar log reductions of 0.74 and 0.71, respectively, but were found to be the least effective of the disinfectants investigated.
A significant difference (P < 0.05) was observed between the 6% hydrogen peroxide (highest average log reduction) and the 70% v/v ethanol (lowest average log reduction). Non-significant differences were observed between both the alcohols tested (P > 0.05). This was also the case with the 6% hydrogen peroxide and the Quat/chlorine dioxide. Both the 6% hydrogen peroxide and the Quat/chlorine dioxide, which was established as being the second most effective of the biocides against the Bacillus spores, were found to differ significantly (P < 0.05) to the amphoteric surfactant, 70% v/v ethanol, 70% IPA, Quat/biguanide, and the SDW. A significant difference was observed between the Quat/biguanide and the Quat/chlorine dioxide (P > 0.05). The log reductions achieved by the amphoteric surfactant also differed significantly from the Quat/biguanide, 70% IPA, and the SDW. Furthermore, there was a significant difference between the 70% ethanol and the SDW (P < 0.05). The most effective of the disinfectants tested against the Bacillus spores were the 6% hydrogen peroxide and the Quat/chlorine dioxide because they achieved the highest average log reductions and differed significantly from the amphoteric surfactant, the alcohols, the Quat/chlorine dioxide, and the SDW.
Discussion
The use of disinfection in aseptic processes is essential, with both the disinfectant itself and its method of application contributing to the overall effectiveness. The best method of disinfection was a combination of spraying followed by wiping with a pre-impregnated wipe. The least effective method was spraying alone. Wiping played a vital role in the removal of the Bacillus spores. The alcohols were among the least effective of the disinfectants tested, with the 6% hydrogen peroxide and Quat/chlorine dioxide being the most effective.
The developed testing method enabled quantitative comparisons to be made between the different biocide preparations under simulated conditions. Differences in log reductions were observed when different methods of disinfection as well as different disinfectants were tested for their efficacy against Bacillus spores. This establishes that the test method used was selective to the method of disinfection, the disinfectant itself, and the test organism.
Spraying
Of the three disinfection methods investigated, spraying alone was found to be the least effective against the Bacillus spores, achieving low levels of log reduction (Figures 1–4). It is possible that some of the spores may have had little or no exposure to the disinfectant due to the delivery of the sprays as droplets, resulting in some areas of the test surface remaining uncovered. The level of coverage of the sprays are of considerable importance because, if only 90% of the test surface were covered, even with 100% efficacy over the remaining area, the 10% recovered from the edge would only enable a log 1 reduction to be achieved. So any spores recovered may be due to lack of contact with the disinfectant rather than the possible low efficacy of the disinfectant. However, the reverse could also occur as some of the disinfectants tested saturated the carrier. This may have resulted in some spores being physically released from the test surface.
Some spores were recovered from the underlying surface following spraying with up to 26% of the spores being washed off (Table III), irrespective of the disinfectant used. Therefore the efficacy that was observed for each disinfection method was not entirely accurate as some spores presumed to have been killed were simply washed off the carrier with up to 2.2 × 107 CFU/mL being released. This may explain why reductions in Bacillus spores were observed on treatment with the two alcohol-based sprays even though alcohols are non-sporicidal (6, 7, 14, 18, 19).
Wiping
The use of disinfectant-impregnated wipes was much more effective in spore eradication than spraying alone (Figures 4 and 5). This result was consistent with two previous studies (1, 3).
Overall, wiping as a method of disinfection was very effective (Figures 1–3). The results achieved on wiping with the disinfectants may be due to the technique of wiping as a physical means of the removal of Bacillus spores, rather than the disinfectant impregnated in the wipe having an effect. This is because all of the biocide wipes tested achieved similar reductions in spore levels and were found to be almost equally effective in spore removal. These data may even be an underestimate of the efficacy of wiping because only the surviving, intact spores could be recovered from the wipes and counted, and any spores killed by the biocide would not be taken into account.
Analysis of the wipes demonstrated that a large number of the Bacillus spores were transferred onto the wipe and not killed by the biocide impregnated within the wipe (Table IV). This was found to be the case with all the wipes tested. However, one disinfectant wipe proved to be better at the physical removal of the spores onto its wipe. This was the wipe with 70% v/v IPA in WFI. Although not expected, the results with this wipe were better than 70% ethanol alone. These results were not statistically significant but show that the composition of the wipe material may play an important role in spore reduction. The wipe impregnated with 70% IPA in WFI, which performed best, was composed of circular knit polyester, and this difference in texture (other wipes were polyester/cellulose) may be responsible for the improved spore removal. Also, the results observed with the SDW-saturated wipe may have occurred because the wipe was composed of 100% cotton. Wiping with pre-impregnated wipes allows direct contact between the spores and the disinfectant. However, spraying disinfectant onto a wipe may result in some parts of the wipe becoming saturated, but other areas remaining dry. Therefore, spraying disinfectant onto wipes is not recommended compared to the use of pre-impregnated wipes.
The numbers of Bacillus spores transferred onto the wipes differed with different species tested. This may be partially explained by differences in spore coat surface characteristics. Analysis of the spore coat by scanning electron microscopy (25) has shown that the predominant surface features are ridges, although the structures differ between and within species. The spore structure of B. pumilis is similar to B. subtilis spores, but B. pumilis spores are much smaller and in some cases the surface is nearly smooth. (26). The smaller size and reduced presence of ridges of the B. pumilis spores may play a role in the results observed because of the reduced friction between the wipe and the spores.
These results indicate that transfer of viable spores from wipes to equipment could be a source of cross-contamination.
Spraying and Wiping
The combination of spraying and wiping achieved slightly higher log reductions (Figures 1–4) than either spraying or wiping alone. The slight differences in log reductions obtained on use of the combination of spraying and wiping showed no significant advantage over wiping alone. However, both wiping alone and spraying and wiping were significantly better at reducing the number of spores than spraying alone and suspension testing. The results observed were as expected (3), as the length of contact time between the disinfectant and the surface to be disinfected has been found to be of great importance. This study employed a contact time of 2 min, although longer contact times generally achieve a higher reduction in spore levels (27). By carrying out disinfection using spray and wipe, the disinfectant exposure increases allowing a build up of concentration. Therefore some spores which may not have been killed on the application of a spray may be killed on prolonged exposure to the disinfectant when it is applied as a wipe. However, it appears that the physical removal of spores by wiping is an important factor to be taken into account.
Suspension Testing
Suspension testing increased the number of log reductions of B. subtilis subsp. subtilis 168 spores than did the hard surface testing (Figure 1). However, this was not the case with the spores of either B. subtilis ATCC 6633 or B. pumilis ATCC 27142 (Figures 2 and 3). This may be due to the increased coverage between the spores and the disinfectants or, as spores can differ in their susceptibility to biocides (21), the B. subtilis subsp. subtilis 168 spores may have been more susceptible to the disinfectants than the other Bacillus spores. Of the disinfectants tested, the alcohols (70% v/v ethanol and 70% v/v IPA) were not expected to have much effect against the Bacillus spores. This is because their thick spore coat prevents alcohols from penetrating and acting upon the protoplast membrane (21). In this study the alcohols were found to exert some effect. The suspension test using alcohols only achieved low log reductions (Figures 1–3), whereas increased reductions were achieved on wiping alone and on a combination of wiping and spraying with the alcohols. This indicates that the wiping can contribute to the removal of spores.
Biocide Efficacy
Of the biocides tested, both the Quat/chlorine dioxide formulation and 6% hydrogen peroxide worked consistently better than the other biocides against all three spore species. The 6% hydrogen peroxide formulation was the most effective, as it achieved higher than a 1 log reduction in each of the disinfection methods tested on hard-surfaces (Figures 1–3). The Quat/chlorine dioxide also achieved a higher than 1 log reduction against the spores in the suspension testing, wiping, and spraying and wiping. Only on application as a spray did it fail to achieve at least a 1 log reduction. This was as expected because although quaternary ammonium compounds are not sporicidal, both chlorine dioxide and hydrogen peroxide are known to possess sporicidal activity, including against spores of B. subtilis (5, 6, 20, 28–30).
It is interesting that the Quat/biguanide and alcohol-based disinfectants 70% v/v ethanol and 70% IPA in WFI, which are considered sporistatic, were effective against spores in this study. Recently, work by Setlow et al. (19) showed that a reduction in B. subtilis spores can be achieved by treatment with ethanol at 65 °C. Alcohols have been found to inhibit germination (18). As the number of viable spores remaining is determined by germination of spores to CFUs, estimations of alcohol-containing disinfectants by this method could overestimate their biocidal efficacy.
The biocide with the least efficacy against the Bacillus spores was the 70% v/v denatured ethanol. The 70% IPA in WFI also achieved fairly low spore reduction levels in comparison to the other disinfectants tested. As alcohols do not possess sporicidal activity (6, 7), the results obtained re-enforce research that indicates the insufficiency of current disinfection protocols that use alcohol sprays and wipes in the aseptic transfer process. Some reduction observed may have been due to the spores being physically removed by the washing (on spraying) or wiping (on wiping) process. The amphoteric surfactant performed poorly against the spores, with only the 70% v/v ethanol achieving a lower average log reduction (Figure 6).
The Quat/biguanide and the SDW were equally as effective in the removal of spores (Figure 6). What was surprising was that on average the SDW was found to be a more effective disinfectant than the amphoteric surfactant and the two alcohol-based disinfectants (Figure 6), although this difference was not found to be significant. However, the effectiveness of the SDW may be due partially to its application with an inert wipe which was found to achieve high log reductions in spore levels. All the disinfectants were found to attain very similar reductions in spore levels in both the “wiping alone” and “spraying and wiping” combination techniques,
As the highest average number of log reductions observed by any of the disinfectants was only 1.22 logs obtained by the 6% hydrogen peroxide, it can be ascertained that the levels of reduction being achieved by these disinfectants is not very high. Also, the relatively low log reduction levels of spores being recovered from the inoculated, disinfected plates indicates that large numbers of spores remain on the stainless steel plates even following spraying and wiping. This contradicts the results obtained in a previous study which was carried out on the same spores (5). This researcher found no spores present on recovery following the wiping alone and spraying followed by wiping methods. None of the disinfectants tested gave a complete eradication of spores from the inoculated stainless steel carriers. Therefore it can be established that none of the disinfectants tested are 100% effective. So even though some reduction is observed, the required level of reduction to satisfy an aseptic sporicidal process is not obtained because log 3 or higher reductions were not achieved (31). Although 3 log reductions were not achieved by any of the disinfectants tested, the level of spores required to be eliminated should be taken into consideration. If relatively low numbers of spores are present, then the use of a sporicidal disinfectant that has a 1–2 log reduction efficacy is likely to be adequate for the majority of pharmaceutical-grade cleanrooms. This is because the total bioburden, including spores, even for the lowest grade of cleanroom should be less than 200 per m3 (31). Therefore, even though a disinfectant agent may not necessarily achieve the 3 log reductions required, it will have some capacity against spores.
Log reductions were found to differ against the three different Bacillus species tested. Higher log reductions were achieved by the biocides against B. subtilis subsp. subtilis 168 spores than against spores of B. subtilis ATCC 6633. Therefore, the disinfectants had different efficacy against different biological test organisms. No single biocide was the most effective against all three Bacillus spore species. Of the Bacillus spores tested, the B. subtilis subsp. subtilis 168 strain was the most susceptible to the disinfectants investigated. This was followed by the B. pumilis ATCC 27142 spores and B. subtilis ATCC 6633 spores, in that order.
Due to the nature of the biocides tested, the results observed in this study generally corresponded with those expected. The results determine that the 70% IPA in WFI and 70% v/v ethanol disinfectants currently being used in aseptic transfer disinfection are not very effective at killing Bacillus spores. As the majority of operators carry out aseptic transfer by spraying with IMS (70% v/v denatured ethanol), it can be concluded that the levels of disinfection currently being achieved in cleanrooms can be increased with the use of other biocides. From the results of this study it is recommended that the 6% hydrogen peroxide and Quat/chlorine dioxide disinfectant biocides be considered in place of the alcohols for use against Bacillus spores. There are issues of concern with both of these agents. Although no residue of hydrogen peroxide exists, due to its harmless breakdown products of water and oxygen, toxicity issues surround its use as a disinfectant. There are concerns about residues of chlorine dioxide being left on surfaces and subsequently entering into components from packaging or trays. Chlorine sprays also tend to be corrosive and are capable of rusting stainless steel isolators. However, this problem may be overcome by employing a wiping with alcohol stage following initial disinfection with the Quat/chlorine dioxide-formulated disinfectant. The use of an aqueous solution which is not volatile is also undesirable, as it will remain on the surface of the transferred items for some considerable time. This may subsequently be transferred into isolators and could be a possible cause of contamination. Operator toxicology is an important factor to consider, as the regular use of disinfectants containing toxicologically harmful chemicals will put operator health at risk.
As standard disinfection procedure, a combination of spraying and wiping should be employed at least once in the transfer process. The use of a wiping stage that follows a spraying stage is recommended because any spores that may have been washed away on a surface on spraying can be picked up and removed with the use of a wipe. Therefore, it is recommended that a wiping stage during transfer of sterile products be incorporated in hard surface disinfection protocols and that these protocols are followed with the greatest care and diligence.
The most effective aseptic disinfection process against the spores was to spray and wipe items with 6% hydrogen peroxide. Considering all the issues as discussed in this paper, the disinfectants and methods of disinfectant application recommended for disinfection of components associated with aseptic processing are spraying with 6% hydrogen peroxide and wiping with a 6% hydrogen peroxide wipe when transferring items into pass-throughs or air-locks into a cleanroom. For the second stage of transferring the items into the isolator hatch to transfer into the compounding area of the isolator, it is recommended that the items are sprayed and then wiped with IMS prior to any manipulation.
Acknowledgements
This work has been financially supported through the provision of a Ph.D. studentship (Manita Mehmi) by the Engineering and Physical Sciences Research Council EPSRC and Shield Medicare Ltd. (CASE/CNA/06/05).
Footnotes
- © PDA, Inc. 2009
