Cross-contamination by disinfectant towelettes varies by product chemistry and strain

Background Disinfectant products are used frequently on environmental surfaces (e.g. medical equipment, countertops, patient beds) and patient care equipment within healthcare facilities. The purpose of this study was to assess the risk of cross-contamination of Staphylococcus aureus and Pseudomonas aeruginosa during and after disinfection of predetermined surface areas with ready-to-use (RTU) pre-wetted disinfectant towelettes. Methods This study tested six disinfectant towelette products against S. aureus ATCC CRM-6538 and P. aeruginosa strain ATCC-15442 on Formica surfaces. Each disinfectant was evaluated on a hard nonporous surface and efficacy was measured every 0.5 m2 using a modified version of EPA MLB SOP-MB-33 to study the risk of cross-contamination. Results We found that all of the wipes used in this study transferred S. aureus and P. aeruginosa from an inoculated surface to previously uncontaminated surfaces. Disinfectant towelettes with certain chemistries also retained a high level of viable bacteria after disinfection of the surface area. The cross-contamination risk also varied by product chemistry and bacterial strain. Conclusion Disinfectant wipes can cross-contaminate hard nonporous surfaces and retain viable bacterial cells post-disinfection, especially over larger surface areas. This highlights a need to further investigate the risk disinfectant wipes pose during and post-disinfection and guidance on maximum surface areas treated with a single towelette.

with mortality rates of approximately 11% among hospitalized patients suffering from HAIs [3].
Staphylococcus aureus and P. aeruginosa are among the most prevalent etiological agents of HAIs [6]. S. aureus is typically harmless to a healthy individual, but can cause deadly infections, such as septicemia, endocarditis, osteoarticular infections, and pleuropulmonary infections [7]. P. aeruginosa can also cause infections such as cystic fibrosis [8] septicemia and pneumonia that can be fatal for immunocompromised individuals [9].
Hard nonporous environmental surfaces in healthcare facilities harbor pathogens that cause HAIs [10]. S. aureus and P. aeruginosa have been detected from bedside cupboards, bed rails, floors and other hospital equipment [11]. S. aureus and P. aeruginosa persist on these surfaces [12,13], increasing transmission risk resulting in HAIs [14]. Overall, pathogen persistence could be due to sublethal concentrations of disinfectants [15,16], low efficacy levels for some classes of disinfectant wipes [17,18], amount of surface area wiped [18], and the label-use recommendations not being followed [19].
Healthcare personnel rely on disinfectant wipes for environmental surface disinfection [20,21]. Previous studies have focused mainly on the bactericidal efficacy of disinfectants under label and off-label use conditions [20,22]. However, despite widespread use of disinfectants, limited studies [23][24][25] have evaluated the risk of disinfectant towelettes cross-contaminating previously uncontaminated surfaces during the wiping process from an inoculum source. The objectives of this study were to (i) evaluate the risk of disinfectant towelettes transferring S. aureus or P. aeruginosa from an inoculated zone to otherwise not contaminated surfaces and (ii) detect levels of S. aureus or P. aeruginosa on disinfectant towelettes after the wiping process. We hypothesized that during the wiping process, disinfectant towelettes are capable of transferring pathogens to otherwise lowrisk areas of an environmental surface. We also hypothesized that post-disinfection, and following label-defined contact times, towelettes may remain contaminated with viable S. aureus or P. aeruginosa.

Methods
Bacteria, disinfectant towelettes, and surface type used for testing This study tested six disinfectant towelettes ( Table 1) that are commonly used in healthcare facilities for equipment and environmental surface disinfection. The experiment proceeded with Formica (laminate) imitation-granite surface, which was two square meters in length as previously described by Nkemngong et al., 2020 (submitted). Briefly, the Formica board was partitioned into five testing zones: the inoculation zone (izone) and one-half (0.5 m 2 ), one (1.0 m 2 ), one and a half (1.5 m 2 ), and two square meters (2.0 m 2 ) from the i-zone (Fig. 1). The zones for testing (swabbing) were 10 × 10 cm (0.01 m 2 ) in size. The disinfectant towelette itself was also analyzed after the wiping procedure was complete.

Surface preparation, wiping method, and sample collection
The Formica was cleaned and disinfected between trials as previously described by Nkemngong et al., 2020. Briefly, the surface was wetted with a 10% bleach solution, followed by rinsing with sterile deionized water. It was then followed by a standard house-hold neutral agent, containing 0.05% Thymol, to reduce chemical residue on the prepared surface. The neutralizing agent was allowed to sit for its recommended contact time (two minutes) before following with a second rinse with sterile deionized water. Lastly, before surface inoculation, 70% ethanol was applied to the Formica sheet and allowed to air dry.
A total of 50 μL of approximately 5.0 × 10 8 log 10 CFU/ ml in 10 μL aliquots (2.5 × 10 7 CFU) were dispensed unto the i-zone, and the wiping procedure followed a modified version of the EPA SOP MB-33-00 for S. aureus and P. aeruginosa [26]. Once inoculated, the culture was allowed to air-dry before using the towelette to wipe the surface from the i-zone to the 2.0 m 2 mark (Fig. 1). From each RTU disinfectant product, the first two towelettes were discarded and the third towelette was used for testing, to ensure it was fully loaded with disinfectant liquid. The wiping process began at the bottom left corner of the Formica board (below where the board was inoculated) and was wiped evenly in an up-down pattern with consistent speed and pressure from the i-zone to the 2.0 m 2 mark. Once the wiping procedure was complete, the disinfectant was left undisturbed for its label-defined contact time (Table 1). At the end of the contact time, the towelette itself was placed in a sterile stomacher bag with 50 mL of 0.52% neutralizing buffer (BD Difco, Becton, Dickinson and Company, MD, USA). Swab samples were also collected from standard 10 cm × 10 cm (100 cm 2 ) sampling areas within each partitioned zone (i-zone, 0.5m 2 , 1.0 m 2 , 1.5m 2 and 2.0m 2 ). Samples for S. aureus or P. aeruginosa detection were collected using PUR-Blue Swabs (World BioProducts, Libertyville, IL; with 10 mL sterile HiCap neutralizing buffer).

Pathogen detection and enumeration
After the wiping procedure and sample collection, each swab sampler was vortexed for 30s to release bacteria cells from the sponge of the swab sampler into 10 mL sterile neutralizing buffer (World BioProducts, Libertyville, IL). Used towelettes in 50 mL of 0.52% neutralizing buffer (BD Difco, Becton, Dickinson and Company, MD, USA) were stomached at 200 rpm for 5 min to release bacteria cells trapped on the towelettes into the neutralizing buffer. Ten mL aliquots from swab samplers and wipe samples were vacuum-filtered onto sterile filter membranes (0.2 μm pore; Pall Corporation, Port Washington, NY) and plated on tryptic soy agar (TSA; BD Biosciences, San Jose, CA) for S. aureus and Reasoner's 2A agar (R2a; Becton, Dickinson and Company Sparks, MD) for P. aeruginosa. Plates were incubated at 37°C for 24 ± 2 h prior to counting colony forming units (CFU).

Statistical analyses
S. aureus and P. aeruginosa were recovered after wiping five separate test zones of a two-meter square Formica sheet and recovered CFU were log 10 -transformed. The disinfectant towelette was also tested for viable S. aureus and P. aeruginosa CFU post-disinfection of the twometer square surface area. Average log 10 CFU loads were calculated for towelettes and sampled surface areas, which were used to test for statistically significant differences among six disinfectant products. The least squares method of the Proc Glimmix test was used to fit linear models (n = 36, α = 0.05) and interactions amongst disinfectant products, sampled surfaces, and log 10 densities on towelettes. Treating both the product type and surface area as continuous variables throughout the data analysis, Tukey adjustments were used to analyze statistically significant differences between mean log 10 CFU/100 cm 2 counts recovered post-disinfection on surface areas treated. The same procedure was used to test for significant differences among average log 10 CFU on used disinfectant towelettes. All statistical tests were conducted using SAS version 9.4 (SAS institute, Cary, NC). Approximately 2.5 × 10 7 log 10 CFU of S. aureus or P. aeruginosa were spotted onto the i-zone (red dots). The surface was wiped in an up and down pattern as indicated by black outlined arrows from left to right. Light grey squares 10 cm × 10 cm (100 cm 2 ) were sampled to recover potentially cross-contaminated S. aureus or P. aeruginosa

S. aureus detected on previously uncontaminated surfaces post-disinfection with towelettes
Irrespective of disinfectant product, ready-to-use disinfectant towelettes transferred S. aureus to previously uncontaminated surfaces from the i-zone (Figs. 2 & 3). Overall, the towelettes transferred on average, 0.19 ± 0.18 and 0.21 ± 0.21 log 10 CFU/100 cm 2 to the 0.5 m 2 and 1.0 m 2 surface areas, respectively. Disinfectant towelettes also transferred a mean of 0.20 ± 0.19 and 0.27 ± 0.34 log 10 CFU/100 cm 2 from the i-zone to the 1.5 m 2 and 2.0 m 2 surface areas regardless of the product being tested.
Regardless of the sampling zone, there were no statistically significant differences among the log 10 CFU/100 cm 2 detected from previously uncontaminated surfaces when wipes were challenged with S. aureus (P > 0.3499; Figs. 2 & 3). The product type was also not statistically relevant overall (P > 0.0756). Specifically, none of the products transferred statistically significant different log 10  Varying levels of P. aeruginosa were transferred by disinfectant wipes to uncontaminated surfaces P. aeruginosa was transferred from the i-zone to previously uncontaminated low-risk surfaces by ready-to-use disinfectant towelettes (Figs. 4 & 5). When transfer levels for all wipes were averaged, a mean of 0.37 ± 0.33 log 10 CFU/100 cm 2 and 0.27 ± 0.23 log 10 CFU/100 cm 2 were transferred from the i-zone to the 0.5 m 2 and 1.0 m 2 surface areas respectively. From the i-zone to the 1.5 m 2 and 2.0 m 2 surface areas, each towelette transferred an average of 0.31 ± 0.26 and 0.35 ± 0.27 log 10 CFU/100 cm 2 , respectively, onto hard nonporous low risk surfaces regardless of the product type.
The cross-contamination risk presented by disinfectant towelettes varies between S. aureus and P. aeruginosa The surface area wiped and strain type were statistically significant (P < 0.0001). Overall, disinfectant towelettes transferred significantly higher log 10 CFU/100 cm 2 of P. aeruginosa than S. aureus to previously uncontaminated surfaces (P < 0.05) and there were significant differences among the tested products (P < 0.05). For both S. aureus and P. aeruginosa, surfaces wiped with HP3, QA1, and QA3 had significantly lower log 10 CFU/100 cm 2 postdisinfection than HP1 and QA2 (P < 0.05). Similarly, for both S. aureus and P. aeruginosa, the log 10 CFU/100 cm 2 detected from the i-zone for HP1 and HP2 were statistically similar after disinfection (P ≥ 0.05; Figs. 2 & 4).
Viable P. aeruginosa and S. aureus were found on disinfectant towelettes after use Overall, the bacterial log 10 CFU remaining on towelettes after disinfection based on the product type used was statistically significant for S. aureus (P < 0.0053) and P.

Discussion
In this study, we evaluated the cross-contamination risk that may be presented by disinfectant towelettes during and after the wiping process. During the wiping procedure, disinfectant towelettes transferred the test pathogens from the i-zone to uncontaminated surfaces (0.5  , with significant differences between the log 10 CFU/100 cm 2 transferred by the products tested. We also found that post-disinfection, all used towelettes retained some level of S. aureus or P. aeruginosa, with some significant differences among the tested products. Overall, HP3, QA1 and QA3 had higher log 10 reductions than the other products tested.
Used disinfectant wipes are potential cross-contamination agents for S. aureus Overall, the tested disinfectant towelettes transferred S. aureus from the i-zone to previously uncontaminated surfaces. This is similar to findings by Ramm et al. who studied the potential for detergent wipes to transfer S. aureus from a contaminated stainless steel surface onto three other surfaces [27]. Specifically, Ramm et al. found that all seven disinfectant towelettes tested crosscontaminated the sterile stainless steel surfaces [27]. The transfer of pathogens to hard nonporous surfaces as demonstrated by our study is particularly important as pathogens are more easily transferred from hard nonporous surfaces to human hands than from porous surfaces [28].
We found no significant differences among the log 10 CFU/100 cm 2 remaining on uncontaminated surfaces post-disinfection. In a similar study with methicillin resistant S. aureus (MRSA) inoculated on stainless steel discs, Williams et al. did not find significant differences in the log 10 CFU detected post-disinfection with towelettes [29]. While overall there were no significant differences in the log 10 CFU/100 cm 2 of S. aureus transferred among all six products we tested, QA1 transferred significantly lower log 10 CFU/100 cm 2 to the 1.5 m 2 zone than all the other products. This difference may be accounted for by the high alcohol content (55% alcohol) of QA1 (Table 1).
Quaternary ammonium towelettes transfer more P. aeruginosa than hydrogen peroxide towelettes Hydrogen peroxide products transferred significantly less log 10 CFU/100 cm 2 to previously uncontaminated surfaces than QA products. This is similar to previous findings by our group where we demonstrated that hydrogen peroxide disinfectant towelettes were more bactericidal against S. aureus and P. aeruginosa inoculated on 97 mm Formica disc than quaternary alcohol towelettes [18]. This significant difference could be explained by the potential for accelerated HP products to produce hydroxyl free radicals which are generally more bactericidal than quaternary alcohols [20]. Moreover, Edwards et al. reported that the number of bacterial cells transferred from contaminated to uncontaminated surfaces is dependent on the disinfectant active ingredient type loaded unto the towelettes [30]. This further emphasizes the role disinfectant chemistry plays in the cross-contamination levels observed in our study.
Among the QA products, QA1 and QA3 had a significantly higher log 10 reduction within the i-zone and also presented a significantly lower cross-contamination risk to uncontaminated surfaces than QA2. Although QA1 and QA3 have differences in their alcohol contents (55% for QA1 and 14.3% for QA3), they had very similar quaternary ammonium contents (0.25% for QA1 and 0.233% for QA3). The similar levels of quaternary ammonium compounds between QA1 and QA3 may have been optimal enough to complement and enhance the bactericidal efficacy of the alcohol levels in these products. This is specifically important as in 2015, Gerba reported that the efficacy of quaternary ammonium compounds is dependent on the product formulation [31].
P. aeruginosa is a higher cross-contamination risk from towelettes than S. aureus Disinfectant towelettes transferred significantly higher log 10 CFU/100 cm 2 of P. aeruginosa than S. aureus. The lipopolysaccharides (LPS) in the outer membrane of Gram negative bacteria may serve as a barrier offering reduced permeability to some disinfectants as this has been specifically demonstrated with quaternary ammonium compounds [32]. In a 2002 study that evaluated the transfer rate of Gram positive and Gram negative bacteria from hard nonporous fomites to the hands of workers, it was found that Gram positive bacteria transfer rates were higher [28].
Regardless of strain, the cross-contamination risk presented by disinfectant towelettes was highly productdependent as HP3, QA1 and QA3 wipes had significantly lower log 10 CFU/100 cm 2 post-disinfection than HP1 and QA2. The observed differences may be explained by factors as differences in the wipe material type [33], active ingredient class [17], and differences in the amount of liquid released onto the test surfaces [18] during the wiping process. The disinfectant towelette substrate type has been demonstrated to play a major role in the physical removal of pathogens from test surfaces [34]. In a 2013 study that compared the bactericidal efficacy of different towelette substrate types composed of either cellulose, cotton, microfiber and a blend of cotton and cellulose loaded with silver dihydrogen citrate, it was found that substrates with a mix of cellulose and cotton were more bactericidal [33]. This may be the case as substrates with cellulose/cotton blends may absorb more disinfectant liquids than the other substrate types tested. It is also likely that differences in the amount of liquid dispensed from the towelettes account for differences we observed in this study and as previously reported [18].

Used disinfectant wipes are potential reservoirs for recontamination after use
After wiping the Formica sheet, all the disinfectant wipes we tested retained some level of S. aureus or P. aeruginosa. This is similar to findings by Cheng et al. who reported MRSA on disinfectant towelettes after use on bedrails [35]. This suggests that disinfectant towelettes could continue to pose a cross-contamination risk if used on larger surfaces (e.g. two meters square surface used in this study). Moreover, in a 2015 study, Ramm et al. reported that detergent towelettes are designed to efficiently pick up and retain microorganisms but that this did not always occur [27]. This finding is consistent with our study as towelettes retained viable cells after disinfection and while the level of risk varies significantly by product for Pseudomonas, there is some level of risk of crosscontaminate for all the wipes studies. An observational study by Williams et al. reported that environmental staff in healthcare facilities used a single disinfectant towelette on at least five different surfaces during routine cleaning and disinfection [36]. This further emphasizes the risk that some disinfectant towelettes present when they retain high levels of common pathogens causing HAI and are used on multiple surfaces by healthcare personnel. Postdisinfection, wipes were found to have significantly higher mean log 10 CFU of P. aeruginosa than S. aureus. The rod shape of P. aeruginosa [37] may have allowed for a better fit into the perforations on the tested towelettes compared to S. aureus cocci [38]. Our study is limited as the impact of the wipe material type was not specifically evaluated. In addition, we did not study the cross-contamination risk towelettes may present in the presence of soil loads.

Conclusion
Overall, disinfectant towelettes transferred viable CFU of S. aureus and P. aeruginosa from the point of inoculation to uncontaminated surfaces, while retaining viable bacterial loads post-disinfection. The results of this study underscore the risk of spreading S. aureus and P. aeruginosa during and after the wiping process by some products. Considering that bacterial biofilms are more prevalent in nature than planktonic bacteria, conducting similar studies against biofilms on environmental surfaces may provide more insights into the possible crosscontamination risk that disinfectant towelettes present. We also recommend conducting similar studies using other Gram negative and Gram positive bacteria relevant in healthcare settings, as well as with mixed cultures to further understand the likely differences in the crosscontamination risk under relevant healthcare scenarios.