Abstract
Methicillin-resistant Staphylococcus aureus (MRSA) can survive for long periods on inanimate objects, and therefore, environmental surfaces constitute an important reservoir for dissemination. However, there is no standardized method for the detection of MRSA from environmental surfaces. The aim of the present study was to evaluate different screening methods to detect environmental MRSA contamination. A total of 294 samples were obtained from inanimate surfaces at a hospital in Luanda, Angola, and a hospital in São Tomé and Príncipe, by three different methodologies: (1) sterile swabs moistened in saline solution, (2) sterile cotton gauzes moistened in tryptic soy broth, and (3) commercial premoistened sterile sponges (polywipes). After a broth enrichment step, all samples were plated onto tryptic soy agar and chromogenic selective media for S. aureus and for MRSA. The S. aureus isolates were characterized by pulsed-field gel electrophoresis, spa typing, multilocus sequence typing, and SCCmec typing. Comparing the three screening methods, gauzes were the most effective (16 S. aureus out of 98 samples; 16.3%), followed by polywipes (4/98; 4.1%) and swabs (3/98; 3.1%). Moistened gauzes were the most sensitive method (p < 0.00001), while screening with swabs was the least efficient (p = 0.00002). The majority of the MRSA isolates (75%) belonged to the main clonal types previously found among patients and healthcare workers in the same hospitals: ST5-IVa (n = 7; 44%) and ST88-IVa (n = 5; 31%). The finding of MRSA on environmental surfaces is dependent on the screening methodology. Moistened gauzes followed by a broth enrichment step proved to be the most sensitive methodology compared to polywipes and swabs.
Introduction
M
Environmental surfaces frequently touched by healthcare workers (HCWs) are commonly contaminated in the rooms of patients colonized or infected with MRSA, which promotes the transmission to susceptible patients through direct contact with contaminated surfaces, through the hands or gloves of HCWs, or through the hands of other patients. 3 In the community, it has been shown that public transportation constitutes an important MRSA reservoir, principally in Portugal4,5 and in the United States, 6 with possible transfer of the pathogen to the hands of passengers. 4
Although several studies evaluated the MRSA contamination of healthcare and community environmental surfaces, the screening methodologies used showed a remarkable degree of variability. Moreover, studies comparing different sampling methods for detection of MRSA are scarce and there is no “gold standard” methodology established.
The aim of the present study was to evaluate different screening methods to improve the detection of MRSA on environmental surfaces. In addition, we determined the population structure of S. aureus isolates recovered from environmental surfaces in two African hospitals.
Materials and Methods
Screened environmental surfaces
In 2017, 49 surfaces were screened at a pediatric hospital (350 beds) in Luanda (Hospital Pediátrico David Bernardino—HPDB) in a single day in January and 49 surfaces were screened at a general hospital (440 beds) in São Tomé and Príncipe (Hospital Ayres de Menezes—HAM) in a single day in March.
The two hospitals were chosen for their high MRSA prevalence among patients and HCWs: HPDB (64%) and HAM (23–30%). 7
Screened surfaces at HPDB included those being potentially frequently touched by patients and/or HCWs, that is, bedside rails, mobile IV stands, bedside tables, medical trolleys, doors, door handles, cleaning equipment, oxygen pumps, plastic folders, wardrobes, telephones, ward taps, and patient chairs. Since at HPDB most of the S. aureus isolates were recovered from bedside rails, posterior sampling at HAM only included bedside rails. All samples were collected before any disinfection. Information concerning the disinfection protocols was not available and probably negligible. Samples were recovered from several wards: (1) HAM: orthopedics, pediatrics, medicine, surgery, and burn unit; HPDB: intensive care unit, surgery, pneumology, and intermediary care units I and II.
Sampling and S. aureus identification
All surfaces were screened by three methodologies: (1) sterile Stuart cotton swabs (Deltalab, Barcelona, Spain); (2) gauzes; and (3) commercial premoistened sponges (polywipes) (Medical Wire & Equipment, Wiltshire, UK). The detailed protocol for each method is presented hereunder:
1. Sterile cotton swabs were moistened in saline solution, immediately rubbed several times over the surface to be sampled, and placed into the transport medium. Swabs were kept in transport medium for 5 days and further inoculated in the laboratory in Portugal in 5 ml tryptic soy broth (TSB) (Becton, Dickinson & Co., Franklin Lakes, NJ) for enrichment. 2. Sterile cotton gauzes (two superimposed layers cut in pieces of 15 × 15 cm) were taken from an individual package, moistened in TSB, and rubbed on to the surface to be screened. For rails, the gauze was wrapped around the rail and rubbed using sliding movements. The gauze was crumpled and immediately inserted into sterile bottles with 50 ml TSB. 3. Sterile polywipes were removed from the bag, rubbed on the surface to be sampled, similarly to swabs, using both sides, and immediately introduced into sterile bottles with 50 ml TSB.
Sampling was always performed first with the swab, which was supposedly the less sensitive methodology. Subsequently, half of the samples were screened with the gauze and finally with the polywipe, while the other half was screened second with a polywipe and finally with the gauze. Sampling with the swab covered an area of ∼30 cm2 over the whole surface. Afterward, one half of the surface was rubbed with the gauze and the other half with the polywipe. Sampling was always conducted by the same researcher. A new pair of sterile latex gloves was used before each sample collection.
Swabs, gauzes, and polywipes placed in TSB for enrichment were incubated at 37°C overnight. The enrichment step was used to increase the sensibility of the three methods in case the bacterial load would be low. All samples were processed in Portugal by the same operator, using the same methodology. A total of 294 samples were plated (100 μl) onto Tryptic Soy Agar (Becton, Dickinson & Co.) and chromogenic selective media for S. aureus (CHROMagar Staph aureus) and for MRSA (CHROMagar MRSA) (CHROMagar, Paris, France). Strains COL (MRSA) and ATCC12228 (Staphylococcus epidermidis) were used as positive and negative controls, respectively, in the selective media. MRSA was confirmed by PCR amplification of the spa gene for species identification and the detection of the mecA gene. 8
Molecular characterization
The isolates were characterized by pulsed-field gel electrophoresis (PFGE), spa typing, multilocus sequence typing, SCCmec typing, and by detection of the Panton–Valentine leukocidin (PVL) gene as previously described. 7
Results
Detection of S. aureus/MRSA isolates
Among the 49 surfaces screened at HPDB, 29% (14/49) were contaminated with S. aureus, out of which 24% (12/49) were MRSA. In HAM, S. aureus was found in 18% (9/49) of the surfaces, and 8% (4/49) were MRSA.
Comparing the three screening methods, most of the S. aureus-positive cultures were recovered from gauzes (16.3%; 16/98), followed by polywipes (4.1%; 4/98) and then swabs (3.1%; 3/98). None of the contaminated surfaces was positive by the three methodologies. In two cases, positive cultures were obtained by both the gauze and polywipe. Moistened gauzes were the most sensitive method for the detection of S. aureus and MRSA (p < 0.00001), while screening with swabs was less efficient (p = 0.00002).
MRSA was recovered from bedside rails (n = 11), cleaning equipment (n = 2), door knobs (n = 2), and a plastic folder (n = 1). Methicillin-susceptible S. aureus (MSSA) was recovered from bedside rails (n = 5), cleaning equipment (n = 1), and a telephone (n = 1).
Characterization of S. aureus isolates
The 16 MRSA and the 7 MSSA isolates belonged to 6 (ST5-IVa, ST88-IVa, ST72-V, ST140-IVg, ST8-V, and ST8-VII) and 5 (ST508, ST1, ST152, ST2446, and ST6) clonal types, respectively (Table 1). The major MRSA clones were PFGE A-ST5-IVa associated to spa type t105 (58%; 7/12) at HPDB and PFGE B-ST88-IVa-t186/t786 (75%; 3/4) at HAM. Two MSSA isolates (ST152) carried PVL.
HAM, Hospital Ayres de Menezes, São Tomé and Príncipe; HPDB, Hospital Pediátrico David Bernardino, Luanda, Angola; PFGE, pulsed-field gel electrophoresis; MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicillin-susceptible S. aureus; PVL, Panton–Valentine leukocidin; ST, sequence type.
One isolate was PVL negative.
Discussion
Since there is a paucity of studies comparing different sampling methods to detect MRSA on environmental surfaces and because of the lack of a standard protocol, we evaluated three different methodologies (swabs, gauzes, and polywipes) on clinical environmental surfaces.
In 2015, in a first study in a hospital in Luanda, we sampled a variety of environmental surfaces with swabs and did not detect any S. aureus (results not shown). Since the MRSA prevalence in that hospital was extremely high (>60%), 7 we hypothesized that the sampling method used was not sufficiently sensitive to detect either MRSA or MSSA. Therefore, in 2017, we performed the present study in the same hospital in Luanda and in another one in São Tomé and Príncipe using not only swabs but also gauzes and commercial polywipes. Most of the MRSA detected were recovered from moistened gauzes, while sampling with swabs was the less sensitive method.
Claro et al. 9 evaluated three methods (Petrifilms, contact plates, and swabs) to assess hospital surface contamination by MRSA and also found that the poorest recovery of MRSA was with swabs. Similarly, studies that used swabs to explore contamination on environmental surfaces in public transportation failed to detect MRSA (in Serbia and France)10,11 or found a very low prevalence (in China). 12 Swabs only make contact with a relatively small area of the surface sampled limiting the transfer of bacteria to the swab. In addition, the recovery rate with swabs may be reduced because the release of bacteria from the swab is incomplete when it is immersed into broth after sampling. 13 This may be an issue for screening environmental surfaces, namely in the community, where the bacterial load may be low. However, the use of flocked swabs could have enhanced the sensitivity of the method, as previously reported when compared with conventional swabs. 13
Although environment sampling using contact plate is recommended for surfaces in high-risk controlled areas, enabling the detection and enumeration of microorganisms, it is only suitable for flat surfaces 9 and therefore not adequate in the present study. While Petrifilms are advantageous over contact plates on the flexibility to adjust to nonflat surfaces, none of the methods allows rubbing of surfaces but just a single contact, which may be not suitable in case of a low inoculum.
Noteworthy, when we screened public buses in Portugal with moistened cotton gauzes (same protocol as in the present study), we detected a massive prevalence of MRSA in vehicles in Oporto (26%) 5 as well as in Lisbon (36%). 4 Moistened gauzes were also used to evaluate the MRSA contamination in bedside rails in rooms of inpatients from whom MRSA was isolated, reporting an MRSA contamination of 32%. 14 The enhanced recovery of MRSA from environmental surfaces with gauzes may be explained by the improved malleability and better adaptation to undulations on the surfaces compared to swabs and polywipes. Moreover, the mechanical friction used with moistened gauzes that perfectly adapt to the hand is superior compared to swabs and polywipes (moistened sponge material does not slip as well), increasing the sensibility of the former method.
Some authors used a neutralizing media to prevent culture inhibition by residual traces of biocide followed by filtration and incubation of the filter on selective agar plates. 15 In our study, disinfection of environment surfaces is not routinely performed in the participating African hospitals and samples were collected before any disinfection. Moreover, eventual residual traces of biocide would be negligible when the specimen was incubated in the enrichment broth.
Characterization of S. aureus isolates showed that MRSA contaminating the environment in the two African hospitals corresponded to the major clones colonizing patients and HCWs in the same healthcare institutions,7,16,17 confirming environmental surfaces play an important role in the dissemination of MRSA in hospitals.
The study presents some experimental limitations. (1) The order on how the samples were collected might have had an impact on the sensitivity of each method. To minimize this problem, we always started by screening with the swab, since we estimated it would be the less sensitive methodology, and for the second and third methods we alternated between sponges and gauzes. (2) It was impossible to screen the same surface with the three methods without removing partially the bacterial load with the previous method(s). To try to overcome this situation, we rubbed one half of the surface with the gauze and the other half with the sponge; however, we could not guarantee the contamination was the same in the two halves of the surfaces. (3) The fact that there was a delay between the culture of the swabs compared with gauzes and polywipes might have decreased the sensitivity of the former methodology. Nevertheless, swabs were kept in transport media, which should maintain the viability of the microorganisms.
In conclusion, the detection of MRSA on environmental surfaces is dependent on the screening methodology. We proposed here a detailed protocol using moistened gauzes, which are malleable and perfectly match the shape of all surfaces, followed by a broth enrichment step that has proved to be the most sensitive methodology for the detection of MRSA compared to polywipes and gauzes. However, the results should be interpreted with caution since in vivo the amount of contamination in the surfaces screened by different methodologies cannot be assumed to be the same. A similar study should be performed in the laboratory, with surfaces contaminated with identical inoculums of MRSA, to corroborate the superior sensitivity of gauzes. Guidelines and standardized protocols are needed for the detection of contamination of environmental surfaces keeping in mind that disinfection of the surfaces is essential in healthcare institutions.
Footnotes
Acknowledgments
We are grateful to the healthcare workers from Hospital Pediátrico David Bernardino and Hospital Ayres de Menezes. We thank Isabel Santos Silva for technical assistance during sampling. This work was partly supported by project PTDC/DTP-EPI/0842/2014 from Fundação para a Ciência e a Tecnologia (FCT), Portugal, by Project LISBOA-01-0145-FEDER-007660 (Microbiologia Molecular, Estrutural e Celular) funded by FEDER funds through COMPETE2020—Programa Operacional Competitividade e Internacionalização (POCI), and by national funds through FCT. Teresa Conceição and Suzilaine Rodrigues were supported by grants SFRH/BPD/72422/2010 and 01/BI/2017, respectively, from FCT, Portugal.
Disclosure Statement
The authors declare no conflicts of interest.