Bacterial survival on inanimate surfaces: a field study

Environmental surfaces may serve as potential reservoirs for nosocomial pathogens and facilitate transmissions via contact depending on its tenacity. This study provides data on survival kinetics of the most important nosocomial bacteria on a panel of commonly used surfaces. Type strains of S. aureus, K. pneumoniae, P. aeruginosa, A. baumannii, S. marcescens, E. faecium, E. coli, and E. cloacae were suspended in 0.9% NaCl solution at a McFarland of 1 and got then plated via cotton swabs either on glass, polyvinyl chloride, stainless steel, or aluminum. Surfaces were stored at regular ambient temperature and humidity to simulate routine daycare conditions. Sampling was performed by contact plates for a time period of four weeks. The longest survival was observed for A. baumannii and E. faecium on all materials (at least four weeks). S. aureus remained viable for at least one week. Gram negative species other than A. baumannii were usually inactivated in less than two days. Nosocomial transmission of the above mentioned bacteria may easily occur if no appropriate infection control measures are applied on a regular daily basis. This might be of particular importance when dealing with outbreaks of A. baumannii and E. faecium.


Introduction
Frequently touched environmental surfaces are described as a major factor of nosocomial transmission [1,2] and the probability of nosocomial spread in those events may be influenced by the tenacity of the particular type of microorganism. Bacteria may highly differ in their potential to survive on such surfaces, but up to now there are only few data available on this topic.
There are some reports on estimations of survival times, but those vary extensively with respect to the inoculum, ambient conditions, and the mode of sampling [3]. So for a better understanding of the true risk of nosocomial transmission, there is a need to better characterize bacteria with respect to environmental survival in a more standardized matter.
This study was carried out to determine the capability of those most relevant nosocomial bacteria to persist over a prolonged period of time on various surface materials. Bacterial suspensions were prepared for each of those eight test organisms from fresh overnight cultures at 37 °C under standard conditions on Columbia 5% sheep blood agar (Becton Dickinson GmbH, Heidelberg, Germany). Colonies from the agar were transferred to the liquid suspension until a McFarland turbidity of 1.0 was reached. Bacteria were suspended in 0.9% NaCl solution in order to avoid potential toxic components that may lead to an accidental primary inactivation. In pre-experiments this amount of microorganisms proved sufficient for growing as a bacterial lawn on contact plates used immediately after plating the suspension.

Surfaces
Survival of the bacteria was tested on glass, polyvinyl chloride (PVC), stainless steel, and aluminum as these materials are frequently used as surfaces in the hospital setting. PVC and other plastic materials are commonly found in form of light switches, shelf spaces for patients, cupboards in bathrooms, bed rails and alarm buttons at the patient´s site. Aluminum may be use for manufacturing hand rails or buttons of elevators. Stainless steel surfaces are very common in doorknobs and levers or in surfaces for the preparation of intravenous infusions or disposal of excretions. Glass surfaces are found on tablet PCs, mobile phones and other touch screens.
Surfaces were thoroughly decontaminated using 70 Vol-% ethanol directly prior usage. For artificial surface contamination, a volume of 25 µL of the bacterial suspensions circulated by pre-soaked cotton swabs was used per spot to ensure that the entire volume remained on the surface. Ten spots per species and surface were prepared for multiple sampling options at different time points (Fig. 1). Surfaces were stored uncovered on the top of wall cupboards at room temperature (21 °C) at a relative humidity of 31 to 35% in order to maintain conditions as given in the routine daycare of patients on a hospital ward.

Sampling
Replicate Organism Dectection And Counting (RODAC; Oxoid Deutschland GmbH, Wesel, Germany) contact plates with a contact surface of 25 cm 2 each were used for sampling over a total period of four weeks. Sampling was primary performed immediately after plating and complete drying of the suspension (day 0) and thereafter on day 1, day 1.5, day 2, day 2.5, day 3, day 7, day 14, day 21, and day 28. Contact plates were then incubated overnight at 37 °C.

Evaluation
The number of recovered colony-forming units (CFU) was determined visually on each plate. If necessary, subcultures of colonies were prepared on an additional Columbia 5% sheep blood agar in order to differentiate between relevant species and environmental contaminants. The experiment was independently carried out thrice (overall 960 samples) and the mean number of CFU from each sampling spot was calculated. For a conservative calculation of the survival time, a value of only 250 was used for further calculation whenever observing a bacterial lawn (uncountable number of CFU). Figure 2 shows the survival kinetics of the test organisms on the four different types of surfaces. Note that A. baumannii and E. faecium showed the highest survival capability regardless of the material of the surface. Viable bacteria of those two species remained detectable even at the end of the entire observation time period of one month. In contrast, survival of all other species was limited to a few days only. However, there were also differences within this rather short surviving panel of species. Gram negative bacteria other than A. baumannii presented with shortest survival times, e.g. P. aeruginosa was completely inactivated in less than two days, while S. aureus remained viable for at least a week on all surface materials tested.

Discussion
Obviously, the length of bacterial survival in the environment impacts the risk of spread. The corresponding time frame depends on multiple factors among them the bacterial species [5] and overall bioburden [6,7], the source of isolation [5], the type of surface material [8,9], the ambient temperature [8,[10][11][12][13], the extent of UV radiation [14], the local pH [13], the relative air humidity [8,11], the availability of water and nutrients [8], the presence of chemical noxa [15], the company by additional (concurrent) bacterial species [11] and other factors like pigmentation [16], and biofilm formation [17]. Table 1 provides a summary of studies on survival times of bacteria in vitro under various conditions. However, most of the results from such previous experiments rely on a rather artificial environment setting, while the study at hand determined the tenacity of nosocomially highly relevant species under conditions as existent in routine daycare of patients. Doing so, we could show that especially A. baumannii and E. faecium are prone for environmental spread in the hospital. This is of importance as antibiotic resistant strains of those two particular species were recently classified as high priority (E. faecium) of even critical priority (A. baumannii) for health-care settings by the WHO [18]. Long-term transmission via environmental contamination in the endemic setting and several outbreaks caused by A. baumannii [19][20][21][22] and E. faecium [23][24][25] are extensively described in the medical literature. Furthermore, D'Sousa et al. identified that A. baumannii and E. faecium even establish synergistic biofilms in vitro when co-cultured [26], which increases the likelihood of prolonged persistence and will facilitate further spread. Thus, our findings confirm the importance of proper infection control measures with emphasis on surface disinfection and/or decontamination procedures.
In recent years there were innovative attempts to reduce the bacterial burden on frequently touched surfaces in hospitals, for example by coating them with layers containing direct bactericide substances or chemicals that diminish biofilm formation [27][28][29]. Another rather novel sanitation strategy is the use of (non-pathogenic) probiotic bacteria that are capable of reducing in a stable way the surface load of pathogens [30] or the use of UV-C light for surface decontamination [14]. However, all of those approaches are still far from comprehensive use in hospitals worldwide so the significance of traditional cleaning and surface disinfection measures will most likely continue for decades.

Generalization of results
Obviously, there are some limitations to our study that need to be addressed. First of all we only tested one single strain of each species. Therefore generalization of our findings should be done with caution. However, Jawad et al. compared the survival times for a total of 39 A. baumannii isolates (22 strains from nosocomial outbreaks and 17 sporadic strains). Their results in terms of survival time were comparable to our findings, but they failed to observe statistically significant inter-lineage differences with respect to bacterial tenacity (26.5 vs. 27.2 days) [31]. On the other hand, there is some newer data suggesting that hydrophilic clonal lineages of A. baumannii possess thicker cell walls and, thus exhibited higher resistance to desiccation compared to hydrophobic strains. This could provide an advantage in environmental survival [32]. Drying resistance of A. baumannii may also depend on mutations and expression of the two-component response regulator gene bfmR, which is important for its virulence and also for the expression of stress-related proteins during a stationary phase [33]. This topic needs to be examined for A. baumannii and the other species alike in more detail in future studies.

Biofilm formation
Secondly, we did not check for the degree of biofilm formation although this may also influence the ability to survive on an inanimate surface [34]. For example, A. baumannii may form strong biofilms on stainless steel surfaces and bacteria within this biofilm are significantly more resistant to environmental noxa than are their planktonic counterparts [35]. E. faecium may also develop biofilms regardless of a concomitant drug resistance but more often in the presence of the esp gene [36][37][38][39]. Ghaziasgar et al. observed this ability even significantly more often in nosocomial isolates while it was less common in wild type strains outside the hospital (100% vs 75.6%; p < 0.05) [40].

Adaptation and virulence of pathogens
Finally, we only measured the number of recovered bacteria via contact plates. Thus, we do not know whether or not changes in the virulence of a pathogen occurred. Although such a phenomenon would not directly affect the transmissibility, it would still be of clinical relevance. Chapartegui-Gonzalez et al. tested five clinical isolates of A. baumannii in long-time survival experiments under simulated hospital conditions. All strains were able to rapidly adapt to both the temperature shift and nutrients availability and maintained their virulence factors despite starvation and desiccation [41]. Once again, similar circumstances apply for enterococci, too [42]. We therefore assume that there was no significant reduction of virulence in the strains used in our study.

Reduction of bioburden by regular decontamination of surfaces
If performed properly, a thorough cleaning and disinfection will significantly reduce the risk of pathogen spread regardless of its tenacity. Unfortunately, breaks in the correct cleaning process are commonly observed due to various reasons. Furthermore small damages to surfaces may cause tiny notches that are then difficult to decontaminate. That is why there are several outbreaks caused by insufficient surface decontamination available in the medical literature. Therefore, this study once again stresses the importance of thorough and regular decontamination of frequently touches surfaces in the hospital for the sake of the safety of patients.