Survival of Vancomycin-Resistant and Vancomycin-Susceptible Enterococci on Dry Surfaces

Abstract

We compared the abilities of Enterococcus faecium strains (three vancomycin-resistant enterococci [VRE] and five vancomycin-susceptible enterococci [VSE]) and Enterococcus faecalis strains (one VRE and 10 VSE) to survive under dry conditions. Bacterial suspensions of the strains were inoculated onto polyvinyl chloride and stored under defined conditions for up to 16 weeks. All strains survived for at least 1 week, and two strains survived for 4 months. A statistical model was used to distribute the 19 resulting survival curves between two types of survival curves. The type of survival curve was not associated with the species (E. faecalis versus E. faecium), the source of isolation (patient versus environment), or the susceptibility to vancomycin (VRE versus VSE). Resistance to dry conditions may promote the transmissibility of a strain, but VRE have no advantages over VSE with respect to their ability to survive under dry conditions.

Since the first reported infection due to vancomycin-resistant enterococci (VRE) (10) many outbreaks of VRE infections or colonization have been reported (3, 4, 7–9, 11–14, 17, 18, 20). At least two of these outbreaks were traceable to environmental sources: one to contaminated rectal thermometers (12) and one to contaminated nursing bells (11).

According to earlier experiences with Staphylococcus aureus, resistant strains may survive longer under dry conditions than susceptible strains (19). These findings raised questions as to whether VRE can survive longer under dry conditions than vancomycin-susceptible enterococci (VSE). We therefore compared the abilities to survive under dry conditions of eight E. faecium strains (three VRE and five VSE) and 11 E. faecalis strains (one VRE and 10 VSE) (Table 1).

TABLE 1.

Strains investigated and sources

Strain	Source of isolation	MIC (μg/ml)		Type
		Vanco-mycin	Teico-planin
E. faecalis 3	Patient (wound)	4	<0.5	VSE
E. faecalis 634	Patient (vascular catheter)	1	<0.5	VSE
E. faecalis 695	Patient (vascular catheter)	1	<0.5	VSE
E. faecalis 4	Patient (skin)	1	<0.5	VSE
E. faecalis O23	Patient (rectal swab)	1	<0.5	VSE
E. faecalis 687	Patient (bronchial lavage)	1	<0.5	VSE
E. faecalis 54	Patient (rectal swab)	16	<0.5	VRE
E. faecalis Z1	Environment (sewage)	1	<0.5	VSE
E. faecalis U6	Environment (litter)	1	<0.5	VSE
E. faecalis U3	Environment (bathroom)	1	<0.5	VSE
E. faecalis ATCC 19433	Unknown	1	<0.5	VSE
E. faecium 547	Patient (wound)	1	<0.5	VSE
E. faecium 26	Patient (urine)	<0.25	<0.5	VSE
E. faecium O1	Patient (rectal swab)	4	<0.5	VSE
E. faecium 36	Patient (urine)	512	128	VRE
E. faecium 166	Patient (rectal swab)	16	<0.5	VRE
E. faecium 1	Patient (ascites)	256	128	VRE
E. faecium Z5	Environment (sewage)	4	<0.5	VSE
E. faecium ATCC 6057	Environment (cheese)	<0.25	<0.5	VSE

The susceptibility of the strains to vancomycin and teicoplanin was tested by the microbroth dilution method. All strains were genetically distinct (as demonstrated by pulsed-field gel electrophoresis, a method described elsewhere [16]).

The strains were stored at −70°C in glycerol broth. They were grown on kanamycin-esculin-azide agar plates at 37 ± 2°C for 24 h.

The bacteria were cultured overnight, washed, and suspended in sterile distilled water. Distilled water was used to rule out a potential protective influence of components of other solutions (e.g., proteins in broth). The final concentrations were determined by a dilution series and averaged 10⁸ CFU/ml.

Polyvinyl chloride (PVC) samples (5 by 5 cm) were disinfected with 70% ethanol and contaminated with 0.1 ml of the bacterial suspension. The total number of contaminated samples was 855 (45 samples per strain). All samples were stored at 22 ± 2°C with 50% ± 5% relative humidity and protected from dust. Viable cells were recovered from the contaminated samples at various time intervals: immediately after drying (zero time) and after 4 h, 1 day, and 1, 2, 4, 8, and 16 weeks. To achieve this, we randomly chose five samples from a total of 45 contaminated samples per strain. These samples were shaken for 5 min in 100 ml of 0.9% sodium chloride solution containing sterile glass beads. The bacterial concentration in the solution was determined in duplicate by membrane filtration and/or a dilution series, and the number of bacteria per sample was calculated.

To evaluate the performance of this method (i.e., the recoverable proportion of bacteria) we chose five samples per VRE strain and tested the bacterial recovery immediately after contamination.

The results were calculated as the average of the bacterial counts per culturing time and strain. The recoverable proportion was calculated as a percentage of the inoculated colony counts. The decrease in the number of bacteria after desiccation was calculated as a reduction factor by using the formula log₁₀ K_o - log₁₀ K_a, where K_o represents the recoverable colony count and K_a represents the colony count after drying. The influences of species, susceptibility to vancomycin, and source of isolation on the recoverable proportion, as well as the reduction after desiccation, were investigated by using a t test for independent samples (SPSS for Windows; SPSS Inc., Chicago, Ill.).

Using the colony counts of the different culturing intervals (zero time to 16 weeks), we drew 19 survival curves. To achieve this, the natural logarithm of the colony counts per strain and culturing interval was expressed as a percentage of the natural logarithm of the colony counts after desiccation (zero time). The resulting survival curves were analyzed by a finite-mixture model and classified into different types (5). Associations between type of survival curve and species, type of survival curve and susceptibility to vancomycin, and type of survival curve and isolation source of the strain were analyzed using the Fisher’s exact test. All tests were performed two tailed and α was set at 0.05.

The recoverable proportion of the strains on the different materials varied greatly, ranging from 8 to 98% (mean, 51%). An association between species and recoverable proportion or susceptibility to vancomycin and recoverable proportion was not found, but strains isolated from patients were recovered in significantly higher proportions than strains isolated from the environment (P = 0.03).

Desiccation reduced the colony counts by 1.2 log₁₀ steps to 6.7 log₁₀ steps (mean, 2.9 log₁₀ steps). There was no significant association between species, susceptibility to vancomycin, or source of isolation and the extent of reduction by desiccation.

All investigated strains survived at least 1 week under dry conditions. Two strains even survived in relatively high colony counts after 16 weeks of investigation (E. faecium 26 [VSE] with 175 CFU/sample and E. faecium 547 [VSE] with 95 CFU/sample). Two further strains were recoverable in low colony counts after 16 weeks (E. faecium 1 [VRE] and E. faecalis 3 [VSE]). Using a finite-mixture model, we separated the 19 survival curves into two types of survival curves (Fig. 1). Type 1 (12 strains) was characterized by early reduction in colony counts. In contrast, type 2 (7 strains) showed later reduction of colony counts.

FIG. 1.

Distribution of 19 survival curves (dotted lines) between two types of survival curves by using a finite-mixture model. Solid bold lines, statistically derived types of survival curves; x axis, survival time in hours; y axis, percentage of natural logarithm of colony count compared to natural logarithm of colony count after desiccation.

We were unable to detect an association between the specific features of the investigated strains (species, susceptibility to vancomycin, or source of isolation) and the type of survival curve (Table 2).

TABLE 2.

Associations between type of survival curve and features of investigated strains

Feature	No. of survival curves		P value
	Type 1	Type 2
Species
E. faecium	4	4	0.38
E. faecalis	3	8
Susceptibility to vancomycin
Susceptible	6	9	1.00
Resistant	1	3
Source of isolation
Patient	6	7	0.59
Environment	1	4

Our results demonstrate that VRE can survive for a prolonged period under dry conditions. This confirms the results of other investigators who found that VRE can survive for 5 days to 2 months on dry surfaces (2, 15); in our study one of the investigated VRE strains even survived for 4 months. However, the ability to survive was not associated with susceptibility to vancomycin. Conversely, the VSE and VRE strains that we investigated were comparable with respect to survival under dry conditions.

Although we delineated two different types of survival curves, we were unable to identify specific features of the strains associated with their ability to survive. For methicillin-resistant S. aureus (MRSA) the pigmentation of the strain and for Acinetobacter baumannii the source of isolation have been related to the ability to survive under dry conditions (1, 21).

The only significant association we found was between the isolation source of the strain and the recoverable proportion. Strains isolated from the environment appear to adhere onto PVC better than strains isolated from patients. This observation should be verified in further studies before reasons and implications can be discussed.

A caveat of our investigation may be that all investigated strains were sporadic isolates. Investigations of MRSA showed that epidemic isolates can survive longer on dry surfaces than sporadic isolates (1). Similar results were obtained for an A. baumannii outbreak strain and two epidemic Burkholderia cepacia strains which survived longer on dry surfaces than sporadic isolates from patients (6, 21). The capability of the microorganisms to survive under dry conditions may lead to better transmissibility and thus to epidemic appearance (6). Further studies should be performed to elucidate whether epidemic VRE excel sporadic isolates in their ability to survive on dry surfaces.

In conclusion, our experiments demonstrated that some enterococcal strains may survive for more than 4 months under dry conditions independent of their susceptibility to vancomycin. Due to heavy contamination with VRE in the environment of VRE patients (3, 7, 9, 12, 22) and the possibility of long survival of VRE, the patients’ inanimate dry environment may be a source for enterococci for a prolonged period even after discharge of VRE patients. To avoid cross-transmission of VRE, cleaning protocols may need to be altered, and especially, cleaning of patient rooms after discharge of the patients should include adequate disinfection procedures.