“High-touch” or frequently touched surfaces may become contaminated with common pathogens of healthcare-associated infections (HCAIs) including methicillin-resistant Staphylococcus aureus (MRSA), Clostridium difficile, vancomycin-resistant enterococci (VRE), and gram-negative organisms such as Escherichia coli, Pseudomonas aeruginosa, Klebsiella pneumoniae, and Acinetobacter baumannii. The risk of cross-contamination between patients is high due to the use of contaminated patient items: mattresses, bed frames, call buttons, over-bed tables, wheelchairs, bedside lockers, etc.Reference Bradbury, Mack, Crofts and Ellison Iii 1 – Reference Hooker, Allen and Gray 4 Moreover, smaller objects (eg, thermometers, blood pressure cuffs, and glucometers) have also been found to have a high bioburden and may be a source of cross-infection between patients.Reference Livshiz-Riven, Borer, Nativ, Eskira and Larson 5
The application of cold-air atmospheric pressure plasma (CAPP) is a nontoxic process. A limited amount of heat is produced (ie, <45°C) in addition to potent radicals (originally from the gas composition) that have antibacterial properties but do not seem to have a destructive impact on surfaces irrespective of composition.Reference O’Connor, Cahill and Galvin 6 Such an approach represents a potentially more attractive alternative to hydrogen peroxide (HP) and ultraviolet light (UVL), which have been increasingly promoted to assist in reducing HCAIs but require the area to be vacated of patients, healthcare staff, and some materials (eg, soft furnishings).Reference Barbut 7 – Reference Galvin, Boyle and Russell 9 We previously documented the efficacy of a single-jet CAPP system to eradicate vegetative bacteria from a variety of surfaces.Reference Galvin, Cahill, O’Connor, Cafolla, Daniels and Humphreys 10
Here, we describe the development and evaluation of a multijet CAPP system to reduce in number a variety of important bacterial pathogens of HCAI from 2 surfaces commonly found in hospitals in the presence and absence of human serum albumin (HSA).
METHODS
Bacterial Strains and Growth Conditions
We chose 2 gram-positive organisms (MRSA and VRE) and 2 gram-negative organisms (E. coli and A. baumannii) for this study. The MRSA strain 43300 and the extended-spectrum β-lactamase–producing E. coli (ESBLEC) strain CL2 are clinical strains from our collection. The VRE strain was provided by the Beaumont Hospital Microbiology Department (Dublin, Ireland) and A. baumannii strain 19606 was obtained from the American Type Culture Collection (ATCC). Before the experiments, bacteria were stored at −20°C in cryovial preservation beads (Microbank, Pro-Lab Diagnostics, Merseyside, UK) and were revived on either Columbia blood agar plates (Oxoid, Basingstoke, UK) for MRSA and A. baumannii, on Mueller-Hinton agar plates (Fluka, Sigma-Aldrich, Ireland) for ESBLEC, or trypticase soy broth agar plates (Oxoid) for VRE. Bacterial cultures were grown aerobically overnight for 16–18 hours at 37°C with rotation in trypticase soy broth supplemented with 5% sodium chloride for MRSA and VRE, brain heart infusion broth for A. baumannii, or Mueller-Hinton broth for the ESBLEC.
Test Surface Preparation
The test surfaces used in this study were 5-cm2 sections of polyurethane mattress (Meditec Medical, Dublin, Ireland) used in hospitals and provided by Beaumont Hospital (Dublin, Ireland) and stainless-steel sections of the same size. Before use, both surface sections were autoclaved and confirmed microbe free. The sections were then aseptically transferred to sterile Petri dishes and placed under UVL for 30 minutes to ensure they were microbe free before use in the experiments.
Preparation of Inocula With and Without Protein
A volume of 25 mL of the appropriate broth was inoculated from an overnight culture plate. Overnight cultures were centrifuged for 10 minutes at 15,500×g (11,000 rpm) (Eppendorf centrifuge 5804R), washed 3 times with phosphate-buffered saline (PBS), and adjusted to a 4 McFarland standard (approximately 9 log10 colony-forming units per milliliter [CFU/mL]) with PBS. For experiments with protein, a 5 McFarland standard (approximately 9.2 log10 CFU/mL) was used, and 300 μL of a 30% in 0.85% sodium chloride solution of HSA (Sigma-Aldrich) was added to 2.7 mL of the McFarland suspension.Reference Speight, Moy and Macken 11 Each test surface was inoculated with a 100-µL suspension spread evenly across the entire surface and allowed to air dry.
Multijet Cold Plasma System
The multijet CAPP prototype (Figure 1) was developed in the National Centre for Plasma Science Technology at Dublin City University. Each plasma jet was driven by a separate sinusoidal, high-voltage power supply. The drive frequency used was 8 kHz, the amplitude of the voltage was approximately 1.5 kV, and each supply consumed an average of 15 W. The operating gas used was dry air, which flowed into the device at 13 L/min. The 9 jets (Figure 1) covered a surface area of approximately 7 cm2 (ie, greater than the test surfaces). The plume-contacting surface temperature did not exceed 45°C, and the distance between the plume and the test surface was approximately 1 cm. The artificially inoculated test surfaces were placed under the jets for periods of 15, 20, 30 and 45 seconds, respectively, to determine the efficacy of the system. All experiments were carried out three times in duplicate (ie, six sets of results).
FIGURE 1 Multijet cold-plasma prototype with electrodes, air inlet, and direction-of-jet array.
Bacterial Recovery and Enumeration
Both test and control (nontreated) surfaces were swabbed using flocked eSwabs (Copan, Italy). Swabs were placed into Falcon round-bottom tubes (BD Bioscience, UK) with 3 mL of PBS, were briefly vortexed, and were cultured onto Columbia blood agar plates for enumeration of MRSA and A. baumannii, ESBL brilliance agar plates (Oxoid) for enumeration of ESBEC and VRE brilliance agar plates (Oxoid) for enumeration of VRE. When needed, 1 and 10 serial dilutions were performed to determine a total viable count (TVC, the number of countable CFUs per milliliter of the sample on the plate [range, 30–300 CFUs]) after plasma treatment. Before exposure to the multijet CAPP, bacterial enumeration was conducted to determine the recovered inocula on the surface as opposed to the applied inocula. Subsequent comparisons in terms of reduced bacterial numbers after plasma exposure were compared to the recovered inocula, not the applied inocula. Statistical analyses were performed using GraphPad Prism 5.00 software (GraphPad, La Jolla, CA). The means of the log values (CFU/mL) from comparisons between recovered control and plasma-treated samples and between samples with and without HSA over 15, 20, 30, and 45 seconds were determined using 1-way analysis of variance (ANOVA).
Atomic Force Microscopy
Atomic force microscopy (AFM) images of control (non–plasma-treated) and plasma-treated bacteria were completed in ambient air with a Dimension 3100 AFM microscope controlled by a Nanoscope IIIa controller (Digital Instruments, Santa Barbara, CA). This equipment was operated in tapping mode using standard silicon cantilevers (Budget Sensors, Nanoworld Holding, Shaffhaussen, Switzerland) with a 7-nm tip radius and a 42-N/m spring constant (nominal values) to assess the physical effects of the plasma on the bacterial cells. Multiple images were examined and edited using WxSM software (Nanotec Electronica, Madrid, Spain) to generate topography, phase, and profile data.Reference Horcas, Fernandez, Gomez-Rodriguez, Colchero, Gomez-Herrero and Baro 12
RESULTS
The effects of the multijet CAPP on A. baumannii, ESBLEC, MRSA, and VRE in the absence and presence of HSA on mattress and stainless-steel surfaces are summarized in Table 1 and are illustrated in Figure 2. The bactericidal effect of the multijet CAPP varied depending on the type of surface, microorganism, and the presence or absence of protein. The most significant effects were observed at the longest exposure time of 45 seconds in the absence of HSA. On the mattress, a 45-second exposure time significantly reduced the bacterial load by approximately log10 3 for A. baumannii (log10 2.92; P<.05) and approximately log10 4 for ESBLEC (log10 4.29), MRSA (log10 4.14), and VRE (log10 3.83). On stainless steel, treatment for 45 seconds reduced the bacterial load of A. baumannii by log10 3.16, ESBLEC by log10 4.69, MRSA by log10 6.21, and VRE by log10 4.36. The addition of protein slightly reduced the efficacy of multijet CAPP, but this reduction was only significant for A. baumannii on the mattress and for VRE on both surfaces (P<.05).
FIGURE 2 Bactericidal effect of the multijet cold-air atmospheric pressure plasma (CAPP) system at 15, 20, 30, and 45 seconds on Acinetobacter baumannii, extended-spectrum β-lactamase–producing Escherichia coli (ESBLEC), methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant enterococci (VRE) in the absence and presence of human serum albumin (HSA).
TABLE 1 Bacterial Log10 Reduction (CFU/mL)±SEM (n≥3) Following 45 Seconds of Treatment with the Multijet Plasma System in the Absence and Presence of Human Serum Albumin (HSA)
NOTE. CFU, colony-forming units; SEM, standard error of the mean; ESBL, extended-spectrum β-lactamase–producing; MRSA, methicillin-resistant Staphylococcus aureus; VRE, vancomycin-resistant enterococci.
The morphological impact of multijet CAPP was evaluated using AFM. Figure 3a and 3b illustrates untreated MRSA cells. Figure 3a shows a 2D image of the cells. Cells are regular, uniform, and intact, as shown by the 3D image in Figure 3b. However, Figure 3c and 3d illustrates MRSA cells following exposure to the multijet CAPP. The arrows in Figure 3d highlight cells that appear to be depressed, flattened, and deformed, possibly due to cell-membrane disruption and the leaking of cytoplasmic material, which strongly suggests that the multijet CAPP physically disrupts the cells. These effects can also be seen in the AFM images of ESBLEC. Figure 4a and 4b illustrates untreated E. coli cells. The cells can be clearly identified in the image as structurally intact, uniform, and characteristically rod shaped. However, in Figure 4c and 4d, the E. coli cells, following treatment, are swollen and distorted (see arrows). These conditions further suggest that CAPP may induce structural instability in the bacterial cells.
FIGURE 3 Tapping mode atomic-force microscopy (AFM) images illustrating the morphological impact of multijet cold-air atmospheric pressure plasma (CAPP) on methicillin-resistant Staphylococcus aureus (MRSA) cells. (a) and (c) topographical images, (b) and (d) corresponding phase images.
FIGURE 4 Tapping mode atomic-force microscopy (AFM) images illustrating the morphological impact of multijet cold-air atmospheric pressure plasma (CAPP) on Escherichia coli cells. (a) and (c) topographical images, (b) and (d) corresponding phase images.
DISCUSSION
The inanimate environment in acute-care hospitals may become heavily contaminated with pathogens of HCAIs and may act as a reservoir for transmission. This condition may arise (1) via direct contact by the patient with that surface or item of equipment, (2) via the hands of healthcare workers, hospital visitors, or others touching that surface and then the patient, or (3) via air.Reference Otter 13 The pathogens implicated include MRSA, VRE, norovirus, C. difficile, and some gram-negative bacteria. Even with general cleaning or disinfection, residual contamination may remain. For example, approximately 25% of rooms were positive for MRSA despite 4 rounds of cleaning and disinfection in a study of terminal cleaning in a 900-bed community hospital.Reference Manian, Griesenauer and Senkel 14
In 2014, Rutala and WeberReference Rutala and Weber 15 reviewed the qualities of an ideal disinfectant: fast acting, nontoxic, easy to use, and economical. Decontamination is often suboptimal because many chemical disinfectants are compromised by their impact on surfaces or by the necessity for vigorous application and because cleaning staff may not be well trained or highly motivated. Enhancing decontamination through the use of HP, UVL, or the incorporation of metals such as copper on surfaces have their limitations, including the need to vacate patients from an area while being disinfected and/or the long-term efficacy of a substance incorporated into a surface.Reference Barbut 7 , Reference Doll, Morgan, Anderson and Bearman 8 , Reference O’Gorman and Humphreys 16
Cold-air plasma (ie, with minimal thermal effects on surfaces or equipment) is an alternative to these approaches and works through the production of reactive oxygen species with antimicrobial activity.Reference Stoffels, Sakiyama and Graves 17 , Reference O’Connor, Cahill, Daniels, Galvin and Humphreys 18 Fridman et alReference Fridman, Brooks and Balasubramanian 19 (2007) demonstrated the efficacy of cold-air plasma on a variety of bacteria and yeasts and even complete eradication after 24 hours of exposure.Reference Fridman, Brooks and Balasubramanian 19 Equivalent plasma processes have also been used to eradicate bacterial biofilms in vitro such as those caused by gram-positive bacteria and P. aeruginosa.Reference Alkawareek, Algwari, Gorman, Graham, O’Connell and Gilmore 20 However, challenges remain in adapting cold-air plasma to the clinical setting, and further evaluations are required.
Modifications to the CAPP system have been made to ensure ease of adaptation. First, the unit has been designed so that nonexpert staff may use it. Other advantages include the short treatment time of 15–45 seconds, whereas other systems generally require far longer treatment times of ~10 minutes to cover large surface areas (eg, UVL systems).Reference Weber, Rutala, Anderson, Chen, Sickbert-Bennett and Boyce 21 To minimize any potential adverse effects, the device has an air recirculation system that carries the plasma radicals away from the surface into a destructor unit. In addition, the unit does not require any consumables (eg, liquid chemicals). These design features ensure both ease of use and patient/staff safety while providing effective decontamination. This system could potentially enhance infection prevention and control procedures on wards and thus contribute to reducing the risk and spread of HCAIs.
The mode of action of the CAPP was evaluated using AFM. The destructive effects of plasma systems on bacteria cells are considered a result of the potent nature of the plasma gas, which contains reactive oxygen species including ozone, H2O2 OH-, and reactive nitrogen species.Reference O’Connor, Cahill, Daniels, Galvin and Humphreys 18 Although the exact mode of action is not fully understood, researchers consider the mechanism to be a combination of the initial etching of the surface by the reactive highly charged gas particles, which then allows these species to enter the cell and induces a cascade of effects that ultimately lead to cell death.Reference Mai-Prochnow, Murphy, McLean, Kong and Ostrikov 22
Limitations of our research include the evaluation of a limited number of bacteria and assessments with high-concentration inocula. The experiments were carried out on pure (not mixed) cultures, as occurs in vivo. We assessed the activity of plasma on only 2 surfaces, and we used HSA as a marker for organic material, whereas many body fluids and other substances are more complex. Furthermore, we did not assess the impact of the multijet CAPP on bacteria in a biofilm. However, we assessed 4 important causes of HCAIs in a research design that simulated, in part, the clinical setting, and we compared reductions in bacterial counts to recovered and not applied inocula.
Interventions to reduce HCAIs require sophisticated design and complex statistical analyses to allow for the many variables that impact the rates of infection.Reference Wolkewitz, Barnett, Palomar Martinez, Frank and Schumacher 23 These results justify further evaluation of a user-friendly CAPP prototype in a carefully designed clinical trial to confirm reductions in surface bacterial counts in busy clinical areas as well as beneficial effects on HCAI rates.
ACKNOWLEDGMENTS
We are grateful to Dr Niall O’Connor for technical assistance and to Dr Deirdre Fitzgerald-Hughes for critically reviewing the manuscript.
Financial support: This study was funded by a translational research award from Science Foundation of Ireland and the Irish Health Research Board (grant no. TRA/2010/10).
Potential conflicts of interest: H.H. receives research funding from Pfizer and Astellas and has in recent years received lecturer or consultancy fees from Cepheid. All other authors report no potential conflicts of interest related to this article.
