Assessing a Novel Method to Reduce Anesthesia Machine Contamination: A Prospective, Observational Trial

Background Anesthesia machines are known reservoirs of bacterial species, potentially contributing to healthcare associated infections (HAIs). An inexpensive, disposable, nonpermeable, transparent anesthesia machine wrap (AMW) may reduce microbial contamination of the anesthesia machine. This study quantified the density and diversity of bacterial species found on anesthesia machines after terminal cleaning and between cases during actual anesthesia care to assess the impact of the AMW. We hypothesized reduced bioburden with the use of the AMW. Methods In a prospective, experimental research design, the AMW was used in 11 surgical cases (intervention group) and not used in 11 control surgical cases. Cases were consecutively assigned to general surgical operating rooms. Seven frequently touched and difficult to disinfect “hot spots” were cultured on each machine preceding and following each case. The density and diversity of cultured colony forming units (CFUs) between the covered and uncovered machines were compared using Wilcoxon signed-rank test and Student's t-tests. Results There was a statistically significant reduction in CFU density and diversity when the AMW was employed. Conclusion The protective effect of the AMW during regular anesthetic care provides a reliable and low-cost method to minimize the transmission of pathogens across patients and potentially reduces HAIs.


Introduction
Healthcare associated infections (HAIs) are a major source of preventable illness, with approximately 2 million occurring in US hospitals every year [1] at a cost approaching 4.5 billion dollars [2]. is poses a serious safety concern associated with a signi cant increase in patient morbidity, mortality, and healthcare costs. Anesthesia machines are known to be active reservoirs for pathogens contributing to the burden of HAIs [1][2][3][4]. e transmission of bacterial pathogens via anesthesia machines occurs both during and between patient cases due to high task density, frequent contact with body uids, invasive procedures, and provider errors, such as omission of adequate hand hygiene. All of these risk factors are performed within the small con nes of the anesthesia work area. ese issues have been identi ed as one of the causes of 30-day postoperative surgical site infection, blood stream infection, central line infection, and ventilator acquired pneumonia in patients undergoing surgery [1][2][3][4]. e source of this contamination could be the transfer of organisms from the patient themselves or from workstation equipment reservoirs such as the anesthesia machine [2,4]. e design of the anesthesia machine makes routine disinfection, sterilization, and cleaning di cult, with complete decontamination all but impossible in daily practice. e adjustable pressure limiting (APL) valves, gas owmeters, and the agent vaporizer dials of the anesthesia machine are common reservoirs for enterococci and other pathogenic bacterial species [2][3][4][5][6][7][8][9][10][11]. Institutional cleaning protocols vary widely and are ine ective in eliminating pathogenic contamination, leading some infection control epidemiologists to refer to the "fecal patina in the anesthesia work area" [6]. In the range of operative settings nationwide, there is likely great variability in cleaning practices. Pathogenic microorganisms are known to survive on the anesthesia machine after standardized, routine cleaning, with bacterial burden reduced but not eliminated, even after accelerated cleaning practices are initiated [5][6][7][8]. In simulations, we and others have demonstrated that routine, between-case cleaning is inadequate in removing a uorescent marker serving as a surrogate pathogen, the inoculum persisting as a potential infection source for a subsequently cared-for patient [6][7][8][9]. Current best cleaning practices fail in achieving full decontamination and thus may place subsequent patients at considerable risk of cross contamination.
To reduce the prevalence of pathogens in the operating room, it may be possible to utilize a disposable, engineered barrier that prevents the contamination of the anesthesia machine during surgery. Use of an ultrathin polyvinyl chloride (PVC) transparent anesthesia machine wrap (AMW) is akin to wearing gloves during procedures and disposing of them after each use, though ergonomics and clinical e cacy have not been systematically studied. An AMW is relatively inexpensive, costing approximately $35.00, with price factors depending on the speci c anesthesia machine in a given OR (http://www.anesthesiahygiene. com, Miami, FL). Its use may reduce horizontal transfer of pathogens to the machine, decrease the potential for machine-vectored contamination in ensuing patients, and improve the e cacy of current cleaning methods. e use of an AMW is intended to supplement universal precautions (UP) in the healthcare setting. Healthcare workers are mandated to practice UP to prevent horizontal and vertical transmission of pathogens.
e Occupational Safety and Health Administration (OSHA) describes, as part of a UP program, that, in addition to self-protection, practitioners "use engineering and work practice controls to limit exposure" [12]. e primary aim of this research was to examine the e ectiveness of the AMW in a clinical setting by testing the following hypotheses: (1) in operating rooms where the AMW is used there will be fewer colony forming units (CFUs) on the anesthesia machine compared to machines without the AMW and (2) in rooms where the AMW was used there will be less diversity in microbial species compared to machines without the AMW.

Description of the Anesthesia Machine Wrap (AMW).
e AMW is shown in Figure 1 (http://www.anesthesiahygiene. com, Miami, FL) and was used on Fabius, GS machines (Drager, Telford, PA) for all trials. In an unpublished pilot simulation study, we determined that the use of the AMW did not impede clinical performance and was readily accepted by a diverse group of experienced anesthesia providers. e AMW has strategically located adhesive strips that allow the wrap to securely adhere to the anesthesia machine. It can be tted onto the machine in less than two minutes and can be removed in seconds avoiding personal contamination, much like a surgical drape. It also has three pouches that may be used for equipment, drug vials, or trash.

Study
Procedure. Following institutional review board approval at Virginia Commonwealth University Medical Center, the AMW was applied to the anesthesia machines in conveniently selected operating rooms where general anesthesia was provided to adult patients undergoing open abdominal surgery; there was no change or in uence on surgical or anesthetic management in any case. e AMW was used in 11 surgical cases (intervention group) and absent in 11 other surgical cases (control group) that were consecutively assigned to two general surgical operating rooms on the three days that the trial was conducted. is sample size was determined to be su cient to reliably identify statistically signi cant di erences in both density and diversity of CFUs between the two groups. Based on prior simulated pilot study results, a large e ect size (approximately 1.25 SD units) was assumed, with one-tailed hypothesis tests. A sample size of 10 observations per condition would provide >80% power to detect a signi cant di erence in contamination; a sample size of 11 per condition was obtained.
Anesthesia providers in the control and intervention (AMW) arms of the study were unaware of the study purpose. ose in the intervention (AMW) were only told that their machines were out tted with a wrap and their perceptions of its ease of use would be queried at case end. Microbial cultures were obtained when both control room and intervention room providers were not present.
Based on the ndings of previous research, microbial analysis was performed on samples collected from seven cultured "hot spots" for bacterial contamination [7,13,14]. We operationalized "hot spots" as targeted areas on the anesthesia machine that were likely to be contaminated by providers' clinical care activities and were di cult to clean/disinfect using routine procedures. e seven "hot spots" were as follows: (i) Vaporizer dial (ii) Patient monitor control panel (iii) Mechanical ventilator control knob (iv) Oxygen owmeter control knob (v) Mouse control for electronic medical record (vi) Keyboard for the electronic medical record (vii) APL valve control knob.
Culture samples for each "hot spot" were obtained prior to rst surgical case of the day in the operating rooms, after the anesthesia provider completed the machine check. is provided the baseline for bacterial contamination in each room (control and AMW rooms). Following collection of samples for culture, the AMW was applied on the anesthesia machine in the study arm; no further manipulation or intervention was performed in the "control" room (no AMW application). After the completion of the surgery and once the patient was transported out of the operating room, the AMW was removed from the anesthesia machine, and samples for culture were obtained from the seven sites listed above. is was performed in all of the operating rooms prior to the routine, between-case cleaning of the anesthesia machines in order to measure any intraoperative contamination that occurred as a consequence of caring for the current surgical patient.
Once the anesthesia machine was cleaned in the standard manner by the anesthesia technicians (who were unaware of the study purpose), the machine in the intervention room had a new AMW applied in anticipation of the arrival of the second surgical patient. No alteration from routine practice was conducted on the anesthesia machine in the control arm of the study (no AMW).

Routine Anesthesia Machine Cleaning Protocol.
Cleaning protocols for anesthesia equipment are highly variable from institution to institution, with no formalized set of national guidelines. is study assessed the AMW at an institution where cleaning and disinfecting processes follow a strict protocol by highly trained anesthesia technicians cleaning/disinfecting the anesthesia equipment based on both institutional and anesthesia machine manufacturer standards. OxyCide ™ (EcoLabs, St. Paul, MN) disinfecting chemicals and wipes are used on each machine at case end and at terminal cleaning at the end of the day. Additionally, an ultra violet (UV) light robot is used following terminal cleaning in each operating room at least once a week.

Performing Cultures.
e seven target surfaces (hot spots) on the anesthesia machine in the intervention and control rooms were cultured for microbiological analysis using ESwabs ™ (COPAN Diagnostics Inc., Waltham, MA) by a clinical microbiologist. e tip of a sterile ocked nylon swab was immersed in tween (a nonionic surfactant) then pressed against the wall of the tube to remove excess solution. e target surface was swabbed using a rotating and twisting motion. Each respective swab was immediately placed in a transport system containing one milliliter of liquid Aimes media. Upon arrival to the laboratory, each ESwab was vortexed for approximately two minutes, and 100 microliters of each respective sample was inoculated to sheep blood agar and MacConkey agar (Copan Diagnostics, Inc., Corona, CA). e inoculum was evenly distributed using a cell spreader, and agar plates were incubated at 35°C for 48 hours. Following incubation, the sheep blood agar and MacConkey agar plates were observed for growth. Each colonial morphotype present on the media was identi ed by gross examination, and CFUs were recorded. Organisms were subcultured to SBA for isolation and identi cation. For the purpose of this study, organisms were identi ed based on colonial morphology, Gram stain, and rapid spot tests such as catalase, Staphaurex ™ Plus latex agglutination, pyrrolidonyl arylamidase, Remel ™ Microdase discs, indole, and oxidase ( ermoFisher Scienti cs, Inc., Middletown, VA).

Statistical Analysis.
Wilcoxon signed-rank tests were used to test the e ectiveness of the AMW for reducing the density of microbiological contamination between the cultured CFUs in covered and uncovered conditions. e Wilcoxon signed-rank test is a nonparametric analog of the t-test that accounts for major deviations from normality. To examine di erences in the diversity of the microbacterial contaminants, a standard t-test was used. Di erences in the density and diversity of CFUs were tested at both the global level and for each speci c hot spot. To account for multiple testing, in addition to the nominal p values for each test, we present Bonferroni-adjusted p values and false discovery rate (FDR) q values. Bonferroni-adjusted p values are known to be extremely conservative, thus greatly increasing the likelihood of a type II error. FDR q values are similar to adjusted p values, but quantifying the likelihood of observing a false positive result from an observation equally or more extreme than that in question. erefore, a q value of 0.022 would imply that 2.2% of the observations at least this extreme is expected to be false positives.

Results
e anesthesia machines covered with the AMW had a statistically signi cant reduction in global density of CFUs recorded across all hot spots with fewer CFUs compared to the uncovered machines (means of the total CFUs on the Control and AMW Machines were 108.0 and 29.2, resp.; p nominal � 0.008, p bonferroni � 0.066, q fdr � 0.022). As can be seen in Figure 2(a), there was also a consistent reduction in the number of CFUs across the speci c hot spots with the AMW in situ, with statistically signi cant di erences for the monitor control panel, the oxygen owmeter knob, and the record keeper mouse. ese data indicate that the use of the AMW was strongly associated with a reduced prevalence of microorganisms.
ere was a statistically signi cant decrease in the diversity of CFUs across all hot spots with the covered anesthesia machines, with the exception of the APL valve machines (means of the total number of distinct CFUs on the Control and AMW Machines were 18.7 and 6.7, resp.; p nominal � 0.0004, p bonferroni � 0.003, q fdr � 0.003). Figure 2(b) presents the results for the diversity of CFUs at the individual hot spots. As can be seen, there was a consistent reduction in the number of phenotypically unique CFUs across sites with the AMW in situ, with statistically signi cant di erences for the vaporizer dial, the monitor control panel, the oxygen owmeter knob, and the record keeper mouse. ese data indicate that the use of the AMW was strongly associated with a reduced diversity of microorganisms.
Because several of the ORs were used repeatedly, it is possible to track the trajectory of the density and diversity of the bacterial contaminates over time. Figure 3 graphically displays the microbial characteristics over the course of surgical case progression revealing a decrease or a stabilization of species in the AMW cases and an increase in species and density over the course of a day in the control room. We interpret these data to indicate that the use of the AMW prevented the introduction of new bacterial species to the anesthesia machine as a result of the anesthesia care rendered from one patient to the next. Figure 3(a) demonstrates a decrease in CFUs from one case to another in the machines that had the AMW in place (blue lines). Likely this indicates that the initial burden of CFUs decreased with subsequent cleaning of the machine after each case, and that no additional bioburden was added due to the presence of the AMW. Alternatively, the unwrapped machines generally experienced increased CFUs likely due to the continued proliferation of existing species or the addition of new species resulting from contamination from a patient being cared for. Figure 3(b) reveals an increase in new species that were added to the unwrapped anesthesia machines, suggesting that patients being cared for, with the anesthesia provider as vector, added to the diversity of bioburden despite cleaning of the machine at case end (red lines). e blue lines reveal a general decrease in species diversity when the AMW was employed, suggesting that the machines were protected from inoculation during patient care. One room in the AMW group (blue line)  experienced an increase in new species that might have occurred during between-case cleaning of the machine or from the provider readying the machine for the upcoming case.

Discussion
is is the rst intraoperative evaluation of an AMW using bacteriological assessment during actual clinical care. e study results show that (1) despite state of the art cleaning protocols, the anesthesia machine remains a reservoir of bacterial species; (2) intraoperative use of an AMW results in statistically signi cant reduction in CFUs on the anesthesia machine compared to care delivered without an AMW; and (3) intraoperative use of an AMW results in a statistically signi cant reduction in the introduction of new species of bacteria onto the anesthesia machine compared to care delivered without an AMW. ese results lead to the conclusion that the AMW reduced the density and diversity of microorganisms in the anesthesia workstation. e importance of engaging in UP during anesthesia care cannot be overemphasized due to the certainty of personnel and equipment coming in contact with biologic material and routinely observed suboptimal levels of hand hygiene compliance, with the anesthesia machine acting as an epicenter for intraoperative pathogens [2-5, 11, 15-17]. As depicted in Figure 4, anesthesia care providers are in nearly constant, repetitive contact with the patient (e.g., skin, mucosa, and oral secretions), the anesthesia workstation (e.g., the vaporizer dial, APL valve, and ventilator knob), and then the patient again (e.g., placing intravenous lines, obtaining blood specimens, Each line represents cases in a particular OR. e blue lines depict cases in which the anesthesia machine was covered with the AMW, and the red lines depict cases where an AMW was not employed. In cases where the AMW was used, there was a decrease or plateauing of the density of CFUs, while in those cases without the AMW there was general increase in the density of CFUs. Relatedly, in cases without the AMW, additional, phenotypically distinct species were added to the anesthesia machine, suggesting ongoing contamination with the anesthesia provider as vector. Note that in cases where the AMW was used, there tended to be a reduction or stabilization in bacterial species, with one exception, which suggests contamination possibly during the cleaning or machine preparation. CFUs � colony forming units; AMW � anesthesia machine wrap.  Canadian Journal of Infectious Diseases and Medical Microbiology and intravenous drug administration). Accordingly, the anesthesia workstation is an essential intermediate source of contamination, and the AMW may reduce the proliferation of pathogens across cases. ere has been a lag in identifying the anesthesia provider as a vector in the genesis of operative infection, largely because a blood stream infection, surgical site infection, or a pulmonary infection may not manifest for many hours (or days) after an inoculation occurred. By that time, the anesthesia provider's role may not be considered. Only recently has the biologic plausibility of an anesthesia provider as a vector been identi ed. Emerging evidence con rms the horizontal movement of pathogens as shown in Figure 4. In fact, transmission of bacteria species of all kinds, including vancomycin-resistant enterococci (VRE), MRSA, and other pathogens occurs frequently and within just a few minutes of care delivery in the anesthesia workstation [11,15,16].
is single center study constitutes a small sample size even though it was su ciently powered to detect large di erences in the presence of bacteria as a function of AMW use. We recognize that our study does not reveal the origin and direction of movement of microbes. e clinical outcomes e ect of the density, and diversity of contamination was not measured in the study. Baseline samples were potentially contaminated since they were collected after the machine check procedure by the anesthesia provider. Additionally, some providers may have better hygienic practices than others; we neither monitored nor controlled to best depict real world practice.
While the AMW may not have a place in routine intraoperative care, this report suggests that there are scenarios where the need for enhanced protection may be indicated, for example, when caring for a patient with a known virulent infectious process (e.g., hepatitis, prion disease, MRSA, VRE, CRE, clostridium di cile, HIV, and Ebola) or when the risk of vertical transmission to the next patient of may have catastrophic consequences (e.g., joint arthroplasty, cardiac surgery, and the immunocompromised patient). While UP are an absolute standard of care, lapses in hand hygiene, di culty in decontamination of complex equipment, and the ergonomics of the anesthesia workstation urge that we develop better methods to reduce infectious iatrogenesis.
Achieving thorough hand hygiene and equipment decontamination may be di cult, if not nearly impossible, despite the high contamination potential in the intraoperative period. A systematic approach, in line with OSHA recommending engineering environmental and practice controls, utilizing a disposable equipment barrier seems logical. We studied the e cacy of the AMW at a site where UP, equipment disinfection, and appreciation for the risk of nosocomial infection at the hands of the anesthesia provider are likely optimal in the real world of clinical practice. We suspect that at sites where there is greater variability in these domains there will be greater degrees of microbial contamination. e results from this study do not suggest that the AMW is a panacea for contamination, but rather that it could be used in conjunction with UP to minimize the risk of infection-based complications after surgery and should not provide users with a false sense of cleanliness. In this vein, three points must be made. First, the AMW will not prevent the spread of airborne pathogens that may persist regardless of the use of any type of protective barrier. Second, standard cleaning procedures will be e ective for macroscopic contaminants that are visible to the naked eyes of the cleaning crew, such as major blood spray. ird, caution is necessary when removing the AMW to avoid contaminating the operating room with what was deposited on the AMW during surgery. We expect that the AMW will be e ective in reducing microscopic contaminants that cannot be seen by the naked eye and/or are deposited on textured surfaces that are di cult to clean.
A list of the bacteria that was cultured in for each hot spot is presented in Table 1. It is important to note that while the bacterial cultures we found in the current study were relatively benign, this does not imply that all organisms that may exist in the OR will always be benign. e prevalence of these organisms is relatively rare, and we would not expect to see them in a study of this size. Across a large number of surgeries, however, even relatively rare pathogens will occasionally be observed. e AMW is intended to minimize the transmission of these relevant, malignant pathogens that may cause serious HAIs.

Conclusion
Considering that use of the AMW entails negligible risk, is low cost, and is e cacious in preventing contamination of the anesthesia machine, this study identi es an opportunity to improve patient safety. e bene ts of such a device may

Conflicts of Interest
e authors declare that there are no con icts of interest regarding their participation in the study, nor potential publication of the ndings.