1
Air and surface contamination patterns of meticillin-resistant Staphylococcus aureus on
2
eight acute hospital wards
3
4
Authors: Eilish Creamer, MSc;1 Anna C. Shore, PhD; 2,3 Emily C. Deasy BA (Mod.); 2 Sandra
5
Galvin, PhD; 1 Anthony Dolan, PhD;1 Niamh Walley, MB;1 Seamus McHugh, MD; 4 Deirdre
6
Fitzgerald-Hughes, PhD; 1 Derek J. Sullivan, PhD; 2 Robert Cunney, MD; 5,6 David C. Coleman,
7
ScD; 3 Hilary Humphreys, MD.1,7*
8
1
9
Ireland, Dublin, 2Microbiology Research Unit, Division of Oral Biosciences, School of Dental Science
10
and Dublin Dental University Hospital, University of Dublin, Trinity College Dublin 2, 3Department of
11
Clinical Microbiology, School of Medicine, University of Dublin, Trinity College, St. James’s Hospital,
12
Dublin 8, 4 Department of Surgery, the Royal College of Surgeons in Ireland and Beaumont Hospital,
13
5
14
Hospital, Temple Street, Dublin, 7Department of Microbiology, Beaumont Hospital, Dublin, Ireland.
15
* Corresponding author:
16
17
Hilary Humphreys, Department of Clinical Microbiology, RCSI Education and Research Centre,
Beaumont Hospital, PO Box 9063, Beaumont Road, Dublin 9, Ireland.
18
Tel.: +353 1 8093710/3708 Fax +353 1 8092871.
19
E-mail address: hhumphreys@rcsi.ie
20
Key words: aerial transmission of MRSA, environmental sampling, air sampling, cross-
21
transmission, MRSA, spa typing, DNA microarray profiling
22
Running title: Aerial dispersal of MRSA
Department of Clinical Microbiology, Education and Research Centre, Royal College of Surgeons in
Health Protection Surveillance Centre, Dublin, 6Department of Microbiology, Children’s University
1
23
Abstract
24
BACKGROUND. Methicillin-resistant Staphylococcus aureus (MRSA) can be recovered from
25
hospital air and from environmental surfaces. This poses a potential risk of transmission to
26
patients.
27
OBJECTIVE. To investigate associations between MRSA isolates recovered from air and
28
environmental surfaces with those from patients when undertaking extensive patient and
29
environmental screening.
30
DESIGN. A prospective observational study of patients and their environment in eight wards of
31
a 700-bed tertiary care hospital during 2010 and 2011.
32
METHODS. Sampling of patients, air and surfaces was carried out on all ward bays, with more
33
extended environmental sampling in ward high-dependency bays (HDBs) and at particular times
34
of the day. The genetic relatedness of isolates was determined by DNA microarray profiling and
35
spa typing.
36
RESULTS. MRSA was recovered from 30/706 (4.3%) patients and from 19/132 (14.4%) air
37
samples. On 9/132 (6.8%) occasions both patient and air samples yielded MRSA. In 32 HDBs,
38
MRSA was recovered from 12/161 (7.4%) patients, 8/32 (25%) air samples, and 21/644 (3.3%)
39
environmental surface samples. On 10/132 (7.6%) occasions, MRSA was isolated from air in the
40
absence of MRSA-positive patients. Patient demographic data combined with spa typing and
41
DNA microarray profiling revealed four transmission clusters, where patient and environmental
42
isolates were indistinguishable or deemed to be very closely related.
2
43
CONCLUSIONS. Air sampling yielded MRSA on frequent occasions, especially in HDBs.
44
Environmental and air sampling combined with patient demographic data, spa typing and DNA
45
microarray profiling indicated the presence of clusters that were not otherwise apparent.
46
3
47
Introduction
48
Numerous studies have shown that the hospital environment is frequently contaminated with
49
potential pathogens that pose a risk of cross-transmission to patients.1,2 Methicillin-resistant
50
Staphylococcus aureus (MRSA) can survive for long periods on environmental surfaces, and
51
may be transmitted to patients via healthcare workers hands or the environment.3,4 Studies on
52
MRSA in the environment have mostly related to outbreaks,5 intensive care units (ICUs), 6,7 or
53
isolation rooms with MRSA patients, rather than in routine ward areas over a period of time.8,9
54
There is relatively little information on the dispersal of airborne MRSA and this may be an often
55
underestimated method of transmission that results in clusters or outbreaks. It is not clear what
56
impact MRSA in the environment and in air have on patient acquisition and cross-transmission
57
and many studies involving the sampling of air in the vicinity of patients have been once-off
58
investigations Previous studies have strongly suggested or confirmed the transmission of bacteria
59
by air, especially when carriers of S. aureus have viral infection.10,11 However, the extent to
60
which MRSA may be recovered from the air and how commonly this occurs in an acute hospital
61
outside of an outbreak is unclear.
62
63
The current study describes the pattern of MRSA recovered from air and the patient’s
64
environment on a number of wards where MRSA is endemic over a period of time and
65
investigates patient and environmental links using spa typing and DNA microarray profiling of
66
isolates.
67
68
4
69
Methods
70
Patients. The study was conducted on patients and their environment on eight wards, four
71
general surgical and four general medical in a 700-bed tertiary referral acute care hospital during
72
2010-2011. The study was undertaken as part of a larger study of MRSA that was approved by
73
the Hospital Ethics (Medical Research) Committee. Patient and environmental sampling were
74
conducted on a scheduled basis, and were not in response to an outbreak investigation. Bed
75
occupancy was approximately 100%.
76
but complementary ways. Firstly, sampling was conducted on 132 occasions on eight wards
77
between March 2010 and February 2011. From March to June 2010, eight wards were sampled
78
successively; three wards were sampled once and five twice. From September 2010 to February
79
2011, the patients and the environment of four wards were each sampled for four weeks
80
consecutively. Patient screening involved the taking of swabs from the nose, groin and non-
81
intact skin/wound if, available (n = 706), with one air sample taken in each ward bay. Secondly,
82
on 32 occasions, extended environmental sampling was conducted in the immediate area of
83
patients in two high-dependency bays (HDBs) on eight wards, where more vulnerable and sicker
84
patients are cared for. Patients were screened for MRSA on study wards as previously
85
described.12,13 At- risk patients were those as defined in national guidelines, i.e. patients known
86
to be previously MRSA-positive, patients transferred from another hospital or a long-term care
87
facility, patients with chronic ulcers or urinary catheters and patients who had been hospitalized
88
within the last 18 months1. Environmental sampling involved surface sampling of each patient’s
89
mattress, pillow, bedrail or bedframe and locker. In addition, a settle-plate was placed on each
90
patient’s locker and one air sampling was conducted at the window-ledge of each ward bay.
Sampling was carried out accordingly in three separate
5
91
Thirdly, air sampling was conducted on HDBs at different times over a 24 h period; between;
92
07.30 -09.00 h, 09.30-11.30 h, 14.00-16.00 h, 17.00-19.00 h, and 20.00-22.00 h, but without
93
simultaneous patient sampling. However during this study phase, routine ward screening did not
94
reveal any MRSA-positive patients to be present in the HDBs during these periods of sampling.
95
No area sampled had any form of artificial ventilation. Cleaning was performed daily by the
96
cleaning services, using water and detergent. A hypochlorite, 1000 ppm av. chlorine was used
97
for MRSA and other infected/contaminated patients. Terminal decontamination of the bed and
98
bed area after a patient’s discharge of the bed and bed area was performed by a designated
99
cleaning team.
100
101
Air and environmental surface sampling. Samples of air (1,000 L) were taken using an impact air
102
sampler (AES Chemunex Air Sampler, Dėpartment SAV, Rue Maryse Bastiė-Ker Lann, 35172
103
BRUZ cedex, France) with MRSASelect chromogenic agar (CA) plates (Bio-Rad Life Science
104
Group, France) which was placed on the ledges of outer wall windows in ward bays. CA settle
105
plates were also placed on patients’ lockers for one hour. Neutralising buffer swabs (Technical
106
Services Consultants, Lancashire, UK) were used to sample mattresses, pillows, bedrail or bed
107
frames and patient lockers. An area, 10 cm x 10 cm, of each surface was sampled. In addition,
108
mattresses were assessed by sweeping a CA plate over the surface of the mattress. Most wards
109
consisted of 35 beds, each with one 4-bedded and one 6-bedded HDB, three other 6-bedded bays,
110
one 2-bedded bay and five single rooms (Figure 1a). HDBs mainly comprised of four (wards A)
111
or six-bedded (wards B) ward bays but one ward had a five-bedded bay. The environmental
112
sampling sites were those immediately associated with the patient’s bed area and are shown in
113
Figure 1b. All environmental swabs were enriched in tryptone soy broth with 6% (w/v) NaCl
6
114
(TB05S-100, Cruinn Diagnostics, Ireland) and incubated at 37°C for 18 h. These were
115
subsequently subcultured onto CA and incubated along with settle plates and sweep plates for 24
116
h at 37°C. Presumptive MRSA was confirmed by coagulase and clumping factor production
117
using Staphaurex Plus (Remel, U.K.) and methicillin resistance determination using oxacillin
118
minimum inhibitory concentration (MIC) evaluator strips (Oxoid, U.K.) and detection of
119
resistance with cefoxitin, 30µg disks (Oxoid, U.K.).
120
Molecular typing. spa typing was carried out on one isolate per MRSA-positive patient and/or
121
the patients’ environment in the HDBs using the method described at the SeqNet website and
122
amplimer sequencing was performed by Source BioScience (Guinness Enterprise Centre, Dublin,
123
Ireland). Where two sampling methods (e.g. swab and sweep-plate) were MRSA-positive from
124
mattress sampling, only one MRSA isolate was included in the analysis. DNA microarray
125
profiling was performed using the StaphyType Kit (Alere, Germany) on all except two of the
126
isolates that underwent spa typing; these two isolates were not available for DNA microarray
127
profiling. The StaphyType Kit detects 334 S. aureus gene sequences and alleles including typing
128
and species-specific markers as well as virulence-associated and antimicrobial resistance genes.
129
DNA microarray procedures were performed according to the manufacturer’s instructions.14
130
Statistical analysis. Statistical analysis was performed using Epi Info 6 (version 6.04c; Centers
131
for Disease Control and Prevention, Atlanta, GA) and odds ratios (ODs) were calculated. The
132
Mantel-Haenszel chi square method was used to assess the significance of the difference
133
between proportions.
134
7
135
136
137
Results
138
yielded MRSA (Table 1). Only two of the MRSA positive patients had infections, the remainder
139
were colonized only. . On 9/132 (6.8%) sampling occasions both patient and air samples were
140
MRSA-positive, but on 10/132 (7.6%) occasions air samples yielded MRSA but no MRSA-
141
positive patients were identified. When four wards were sampled consecutively for a four-week
142
period, MRSA was isolated on week 1, (5 patients, 2 air samples); week 2 (4 patients, 2 air
143
samples); week 3, (1 patient, no air samples) and week 4 (2 patients, 1 air sample). Three of the
144
five air samples were linked to a MRSA-positive patient in a specific ward bay.
Patient and air sampling. Overall, 30/706 (4.3%) patients and 19/132 (14.4%) air samples
145
146
Environmental sampling on 32 occasions in HDBs. MRSA was recovered from 9/85 (10.6%)
147
male and 3/76 (4%) female patients on nine environmental sampling occasions resulting in
148
(12/161 (7.4%) positive samples. MRSA was recovered from 8/32 (25%) air samples, from
149
12/161 (7.5%) settle plates and from 21/644 (3.3%) surface environmental sites. On five
150
occasions, MRSA was isolated from two patients in a ward bay. Ten of 161 (6.2%) mattresses,
151
2/161 (1.2%) pillows, 3/161 (1.9%) bedrail/frames and 6/161 (3.7%) bedside lockers yielded
152
MRSA. MRSA was isolated from at least one sample (patient or environment) on 16/32 (50%)
153
occasions. On 2/32 (6.3%) occasions, MRSA was recovered from one or more patients as well as
154
from air samples, settle plates and one or more environmental surface sites. For three of twelve
155
MRSA-positive patients, MRSA was recovered from the patient’s mattress, plus the locker for
156
one patient, from the mattress and locker for another patient, and from the mattress, bed frame
157
and pillow of another patient. Two of these MRSA-positive patients were associated with
8
158
extensive contamination, (ward 2 in June 2010 and ward 7 in January 2011). Patient risk factors
159
for the 12 MRSA patients included previous MRSA (1/12), transfer from another hospital (1/12),
160
the presence of a urinary catheter (3/12) and admission within the previous 18 months (7/12).
161
MRSA was isolated from nasal swabs in ten patients, a groin swab in one and both nasal and
162
groin in one patient. One of 12 patients with nasal/groin swab MRSA-positive was associated
163
with extensive MRSA environmental contamination in January 2011.
164
Air sampling over a 24 h period. Thirteen of 128 (10%) air samples in HDBs yielded MRSA;
165
7/32 (22%) between 07.30-09.00h, 3/32 (9%), between 09.30-11.30h, 1/32 (3%), between 14.00-
166
16.00h, 0/16 (0%), between 17.00-19.00h and 2/16 (13%), between 20.00-22.00h, with
167
significantly more MRSA-positive samples recovered between 07.30h and 09.00h compared to
168
other times, i.e. 7/32 (22%) versus 6/96 (6%), odds ratio (OR), 4.20 (95% confidence interval
169
(CI), 1.13-15.81), p=0.01.
170
Molecular epidemiological typing. spa typing was conducted on 42/53 (79%) MRSA isolates
171
arising from extended patient and environmental screening in HBDs; 10 patient isolates, five air
172
samples, nine settle-plate isolates and 18 environmental surface isolates. Four different spa types
173
were identified among patient and environmental isolates including t515, (19 isolates) t557, (19
174
isolates), t032, (four isolates) and t022 (one isolate). DNA microarray profiling was performed
175
on 42 of the isolates that were spa-typed. The array assigned all isolates to ST22-MRSA-IV and
176
revealed that all harbored the beta-lactamase resistance gene blaZ, the enterotoxin gene cluster
177
egc and the immune evasion complex (IEC) genes sak, chp and scn. Limited variability was seen
178
among the isolates in relation to the carriage of the macrolide resistance gene erm(C) (37/40,
179
92.5%) and the co-located enterotoxin genes sec & sel (36/40, 90%). The DNA microarray
9
180
profiles of all spa type t557 isolates were indistinguishable (blaZ, erm(C), sec & sel, egc and IEC
181
type B, n = 20). Two array profiles were identified among t515 isolates (blaZ, erm(C), sec & sel,
182
egc and IEC B, n = 12; blaZ, erm(C), egc and IEC B, n = 3), whereas three were identified
183
among t032 isolates (blaZ, egc, IEC B, n = 1; blaZ, sec & sel, egc, IEC B, n = 1; blaZ, ermC,
184
sec & sel, egc, IEC B, n = 2).
185
186
On epidemiological grounds there were four potential MRSA clusters, i.e. MRSA isolated from
187
one or more patients and their environment or from two or more patients without any
188
environmental sites positive for MRSA. Patient demographic data combined with spa typing and
189
DNA microarray profiling suggested four clusters (Table 2). The first of these was on the
190
06/11/10 and involved 19 spa type t557 isolates from three patients with 16 environmental sites
191
in two bays of ward 2; these isolates all yielded indistinguishable DNA microarray profiles. The
192
second spa typing cluster was recovered on the 14/06/10 and involved two spa type t032 isolates
193
from two patients in two ward bays of ward 3; these isolates differed by one gene combination
194
only using the DNA microarray and are considered to be closely related. The next spa typing
195
cluster was recovered on 26/10/10 and again involved two spa type t032 isolates, one recovered
196
from a patient and the other from an environmental site of ward bay A on ward 5; these isolates
197
exhibited indistinguishable DNA microarray profiles. The final spa typing cluster involved 12
198
spa type t515 isolates recovered on 01/04/11 and involved one patient and 11 environmental
199
isolates on ward A; again these isolates exhibited indistinguishable DNA microarray profiles.
200
Three of four other environmental isolates yielded indistinguishable spa (t515) and microarray
10
201
pattern profiles, on the 23/06/2011, in two bays of ward 5. However, no positive MRSA patients
202
were identified as being epidemiologically associated with these isolates.
203
Discussion
204
MRSA is endemic in most Irish hospitals and this study was part of a program of research to
205
determine if current methods of screening underestimate the extent of MRSA colonization and
206
environmental contamination. Through a combination of admission screening of all patients (not
207
just those in at-risk categories such as patients being transferred from another hospital), follow-
208
up screening while in-patients, air and environmental sampling and the prospective collection of
209
detailed demographic data and molecular typing, we have endeavored to gain a deeper
210
understanding of the extent of MRSA colonization and transmission. Results showed that some
211
patients are MRSA-positive even though not in an at-risk category and who would not normally
212
be routinely screened, that transmission can occur but remain undetected if solely relying on
213
routine screening and that the method of environmental sampling influences the density of
214
MRSA recovered from the healthcare environment.12, 13, 15
215
216
In this study MRSA was mainly isolated from the air in ward bays, (air sampling sites outlined in
217
Figure 1b), where MRSA-positive patients were identified. However, on 7.6% of 132 sampling
218
occasions, MRSA was detected from air samples when no known MRSA-positive patients were
219
identified, possibly because of unidentified MRSA-positive individuals in the vicinity, recently
220
discharged MRSA-positive patients, or environmental contamination. When patient screening
221
and air sampling were conducted on four wards over a 4-week period, more MRSA-positive
222
patients and environmental samples were identified during the first two weeks, than during
11
223
weeks 3-4. This may suggest that the early identification of MRSA patients arising from this
224
study and their subsequent isolation may have contributed to fewer MRSA-positive samples
225
during the last two weeks of screening.
226
227
In HDBs, MRSA was recovered on 50% of occasions from either patients or their environment,
228
probably due to the greater risk of these more vulnerable patients having MRSA. Apart from the
229
presence of a urinary catheter in 3/12 patients, no other MRSA risk factors were associated with
230
environmental contamination. Skin scales can travel significant distances especially when
231
associated with extensive activity.9,16,17,18 While we assessed particle counts and particle size
232
variation on one occasion (data not shown), we found no correlation with bacteriological
233
sampling results.
234
235
Active air sampling by its very nature draws air from a wide area, which makes it difficult to
236
associate environmental isolates with specific patients. In order to more accurately measure the
237
dispersal of MRSA from individual patients we conducted environmental sampling of beds. For
238
three of twelve MRSA-positive patients, MRSA was recovered from the patients mattress as well
239
as other environmental sites surrounding the patient’s bed. Two of these patients were associated
240
with extensive environmental contamination on ward 2 on the 11/06/2010 and on ward 7 on the
241
04/01/2011. While not conclusive, these patients may have been index patients for the outbreaks.
242
The greater likelihood of recovering MRSA from air samples compared with settle plates and
243
environmental surfaces is possibly related to the increased sensitivity of this method of sampling
12
244
due to the increased volume of air sampled (i.e. 1,000 L).19 Seven of 12 (75%) MRSA-positive
245
settle plates were related to two clusters involving four patients, suggesting that MRSA can be
246
dispersed in the air from the immediate vicinity of MRSA-positive patients to other ward areas.
247
248
Rates of recovery of MRSA are higher during ward activities, and at different times during the
249
day, before, during and after cleaning.18,6 We recovered significantly higher rates of MRSA in
250
air samples early in the morning, i.e. between 07.30-09.00 h compared with at other times.
251
Sampling at this time, before routine ward cleaning would generally be regarded as a low activity
252
time, i.e. less movement of patients and staff.20-22 Air samples positive for MRSA declined
253
during the day when cleaning, ward rounds and other activities were taking place, with less
254
MRSA being isolated during the afternoon/evening, similar to a study on two respiratory wards
255
that showed higher levels of particulate matter in the morning during ward activity , i.e. bed
256
making, nebulisation, peaking again around lunch time and reducing during times of low
257
activity.12 The more frequent recovery of MRSA from surfaces in ICUs has been reported
258
between 08.00-09.00 h than during the rest of the day.6 Aerial dispersal of Clostridium difficile
259
was found to be higher during the distribution of meals.23 However, patients may shed more skin
260
scales6,23 during the night when sleeping, and this may explain our finding of more MRSA
261
between 07.30-09.00 h, but some ward activity may have been taking place before 7.30 AM.
262
263
Effective cleaning and the decontamination of environmental surfaces have been associated with
264
reduced MRSA in the patient’s environment,24 but cleaning may not always be effective. In the
13
265
present study, mattresses were the items of equipment from which MRSA was most frequently
266
isolated and these may be a significant reservoir in the environment. Patients may acquire MRSA
267
from a previous MRSA-positive patient if decontamination between patient admission to that
268
clinical area is inadequate.25 The surface of patients’ lockers was also contaminated, possibly
269
from aerial transmission, but also due to indirect contact with MRSA-positive patients via
270
healthcare staff combined with inadequate cleaning.
271
272
The use of spa typing as a standalone typing method for differentiating between highly clonal
273
ST22-MRSA-IV isolates has been shown to be inadequate.26 In the present study spa typing was
274
combined with DNA microarray profiling to attempt to enhance discrimination of the ST22-
275
MRSA-IV isolates based on the presence or absence of virulence-associated and antimicrobial
276
resistance genes. This combined molecular typing approach together with epidemiological data
277
identified four clusters, three of which were on different wards involving patients and the
278
environment. In addition, three of four isolates that were linked epidemiologically and isolated
279
from the environment were indistinguishable but these were not associated with MRSA-positive
280
patients. While it is important not to rely on spa typing alone for typing of MRSA particularly in
281
an endemic setting, in the present study DNA microarray profiling did not result in isolates
282
clustered based on spa typing being further differentiated. However, the DNA microarray
283
permits identification of characteristic resistance and virulence genes among isolates as we have
284
demonstrated for a collection of isolates causing MRSA bloodstream infection.27
285
14
286
Patterns of air dispersal of S. aureus are not generally well understood in the healthcare
287
environment or elsewhere. A recent study of livestock-associated MRSA traced the spread of
288
MRSA in the vicinity of pig barns and spa typing indicated that MRSA spread was influenced by
289
the season and the wind direction.28 The clinical areas sampled in our study were naturally
290
ventilated and therefore air flow would have been influenced by open doors and windows and
291
the movement of patients and staff.
292
293
The limitations of this study include taking samples mainly in HDBs and not as frequently
294
throughout the rest of the ward, as this would have more accurately determined the full extent of
295
MRSA dispersal. We sampled mainly nasal and groin sites but not other sites such as throat,
296
sputum, urine, etc. which may be associated with significant additional air dispersal. Because of
297
logistical and resource reasons, air sampling was conducted at one position only, i.e. the outer
298
window of each ward bay and not beside each patient in all ward bays, thus our sampling
299
possibly missed MRSA dispersion from patients further away from the window. Sampling was
300
conducted during the day and early night periods, but not between 22.00h-07.30h, when patients
301
were asleep and there was less activity on the ward. Broth enrichment was not used for patient
302
samples and healthcare staff were not screened; both may have underestimated the true number
303
of MRSA- positive carriers on the ward and may explain the presence of MRSA in the air
304
without known MRSA positive patients in the immediate vicinity. One study found that the
305
average limit of detection for direct culture and broth enrichment in an in vitro study was
306
approximately 750 CFU/ml and 40 CFU/ml, respectively.29 Hence broth enrichment significantly
15
307
enhances the sensitivity of MRSA detection but with a longer time to confirmation of MRSA
308
status.
309
310
In conclusion, MRSA was recovered from either patients or their environment on 50% of
311
sampling occasions with patterns suggesting dispersal from patients to the surrounding air. This
312
suggests that aerial transmission from patient-to-patient may be common. While not conclusive,
313
patients appear to shed MRSA more frequently early in the morning or during the night. The use
314
of computational fluid dynamics might assist in the accurate prediction of the spatial distribution
315
of MRSA as shown in a recent study which investigated the use of a physical partition between
316
patients and the location of patients vis a vis the air supply.30 However, further prospective
317
studies of patients in different settings including in single rooms and in multiple bed settings,
318
during different periods of activity such as the changing of bed linen, and under natural and
319
artificially ventilated facilities are required.
320
321
Acknowledgements
322
This study was funded by the Health Research Board, Ireland (TRA/2006/4). We are grateful to
323
all members of the Infection Prevention and Control Team and the Department of Microbiology
324
in Beaumont Hospital for their assistance. We also acknowledge the co-operation,
325
encouragement and support from all nursing, medical and other staff.
16
326
Conflict of interest: HH has received research support from Steris Corporation, Inov8 Science,
327
Pfizer & Cepheid in recent years. He has also recently received lecture & other fees from
328
Novartis, Astellas and AstraZeneca.
329
No conflict of interest is reported by the other co-authors.
330
17
331
References
332
1
Health Protection Surveillance Centre. The control and prevention of
333
MRSA in hospital and in the community. Dublin: Health Protection
334
Surveillance Centre; 2005.
335
2
Smith TC, Moritz ED, Leedom Larson KR, Ferguson DD. The
336
environment as a factor in methicillin-resistant Staphylococcus aureus
337
transmission. Rev Environ Health 2010;25: 121-134.
338
3
Boyce JM, Potter-Bynoe G, Chenevert C, King T. Environmental contamination due to
339
methicillin-resistant Staphylococcus aureus: possible infection control implications.
340
Infect Control Hosp Epidemiol 1997;18:622-627.
341
4
342
343
reservoirs for nosocomial infection? Clin Infect Dis 2004;39:1182-1189.
5
344
345
Hota B. Contamination, disinfection, and cross-colonization: are hospital surfaces
Otter JA, Yezli S, French GL. The role played by contaminated surfaces in the
transmission of nosocomial pathogens. Infect Control Hosp Epidemiol 2011;32:687-699.
6
Wilson AP, Smyth D, Moore G, et al. The impact of enhanced cleaning within the
346
intensive care unit on contamination of the near-patient environment with hospital
347
pathogens: A randomized crossover study in critical care units in two hospitals. Crit Care
348
Med 2011;39:651-658.
349
7
Wilson RD, Huang SJ, McLean AS. The correlation between airborne methicillin-
350
resistant Staphylococcus aureus with the presence of MRSA colonized patients in a
351
general intensive care unit. Anaesth Intensive Care 2004;32: 202-209.
18
352
8
Sexton T, Clarke P, O'Neill E, Dillane T, Humphreys H. Environmental reservoirs of
353
methicillin-resistant Staphylococcus aureus in isolation rooms: correlation with patient
354
isolates and implications for hospital hygiene. J Hosp Infect 2006;62:187-194.9.
355
9
Gehanno JF, Louvel A, Nouvellon, M, Caillard J-F, Pestel-Caron M. Aerial dispersal of
356
meticillin-resistant Staphylococcus aureus in hospital rooms by infected or colonised
357
patients. J Hosp Infect 2009;71:256 -262.
358
10
359
360
Sherertz RJ, Reagan DR, Hampton KD, et al. A cloud adult: The Staphylococcus aureus virus interaction revisited. Ann Intern Med1996;124:539-547.
11
Yap FHY, Gomersall CD, Fung KS, et al. Increase in methicillin-resistant
361
Staphylococcus aureus acquisition rate and change in pathogen pattern associated with an
362
outbreak of severe acute respiratory syndrome. Clin Infect Dis 2004;39: 511-6.
363
12
Creamer E, Galvin S, Dolan A et al. Evaluation of screening risk and non risk patients for
364
methicillin-resistant Staphylococcus aureus on admission in an acute care hospital. Am J
365
Infect Control 2012;40: 411-5.
366
13
Creamer E, Shore AC, Rossney AS, et al. Transmission of endemic ST22-MRSA-IV on
367
four acute hospital wards investigated using a combination of spa, dru and pulsed field
368
gel electrophoresis typing. Eur J Clin Microbiol Infect Dis 2012;31:3151–3161.
369
14
Monecke S, Jatzwauk L, Weber S, Slickers P, Ehricht R. DNA microarray-based
370
genotyping of methicillin-resistant Staphylococcus aureus strains from Eastern Saxony.
371
Clin Microbiol Infect 2008; 14: 534-45.
372
15
Dolan A, Bartlett M, McEntee, B, Creamer E, Humphreys H. Evaluation of different
373
methods to recover meticillin-resistant Staphylococcus aureus from hospital
374
environmental surfaces. J Hosp Infect 2011;79: 227-30.
19
375
16
376
377
Soc. Interface 2009; 6:S697–S702.
17
378
379
Eames I, Tang JW, Li Y, Wilson P. Airborne transmission of disease in hospitals. J. R.
Roberts K, Hathway A, Fletcher LA, Beggs CB, Elliott MW, Sleigh PA. Bioaerosol
production on a respiratory ward. Indoor Built Environment 2006: 15: 35-40.
18
Shiomori T, Miyamoto H, Makishima K, et al. Evaluation of bed making-related airborne
380
and surface methicillin-resistant Staphylococcus aureus contamination. J Hosp Infect
381
2002;50:30-35.
382
19
383
384
Infect 2000;46:241-256.
20
385
386
21
22
Solberg CO. Spread of Staphylococcus aureus in hospitals: causes and prevention. Scand
J Infect Dis 2000;32:587-595.
23
391
392
Greene VW, Bond RG, Michaelsen MS. Air handling systems must be planned to reduce
the spread of infection. Mod Hospit 1960; 95: 136-144.
389
390
Wong V, Staniforth K, Boswell TC. Environmental contamination and airborne microbial
counts: a role for hydroxyl radical disinfection units? J Hosp Infect 2011;78:194-199.
387
388
Pasquarella C, Pitzurra O, Savino A. The index of microbial air contamination. J Hosp
Best EL, Fawley WN, Parnell P,Wilcox MH. The potential for airborne dispersal of
Clostridium difficile from symptomatic patients Clin Infect Dis 2010; 50:1450–1457.
24
Dancer SJ, White LF, Lamb J, Girvan EK, Robertson C. Measuring the effect of
393
enhanced cleaning in a UK hospital: a prospective cross-over study. BMC Medicine 2009,
394
7:28 http://www.biomedcentral.com/1741-7015/7/28 accessed 27-08-2011.
395
396
25
Huang SS, Datta R, Platt R. Risk of acquiring antibiotic-resistant bacteria from prior
room occupants. Arch Intern Med 2006;166:1945-1951.
20
397
26
Shore AC, Rossney AS, Kinnevey PM, et al. Enhanced discrimination of highly clonal
398
ST22-methicillin-resistatn Staphylococcus aureus IV isolates achieved by combining spa,
399
dru, and pulsed field gel electrophoresis typing data. J Clin Microbiol 2010; 48: 1839-52.
400
27
McNicholas S, Shore AC, Coleman DC, Humphreys H, Fitzgerald Hughes D. DNA
401
microarray genotyping and virulence and antimicrobial resistance gene profiling of
402
methicillin-resistant Staphylococcus aureus bloodstream isolates from renal patients. J
403
Clin Microbiol 2011; 49: 4349-4351.
404
28
Schulz J, Friese A, Klees S, et al. Longitudinal study of the contamination of air and soil
405
surfaces in the vicinity of pig barns by livestock-associated methicillin-resistant
406
Staphylococcus aureus. Appl Environ Microbiol 2012; 78:5666-71.
407
29
Rossney AS, Herra C, Brennan GI, Morgan PM, O’Connell B. Evaluation of the Xpert
408
methicillin-resistant Staphylococcus aureus (MRSA) assay using the GeneExpert real-
409
time PCR platform for rapid detection of MRSA from screening specimens. J Clin
410
Microbiol 2008; 46: 3285-3290.
411
30
King MF, Noakes CJ, Sleigh MA, Camargo-Valero MA. Bioaerosol deposition in single
412
and two-bed hospital rooms: a numerical and experimental study. Build Environ 2012;
413
doi: 10.1016/j.buildenv.2012.09.011.
414
415
416
417
418
419
420
21
421
Figure 1a. Schematic layout of ward bays in a typical 35-bed ward
422
423
2-bed 6-bed
6-bed
6-bed
6-bed
4-bed
Bay F Bay E
Bay D
Bay C
Bay B
Bay A
SR
SR
SR
SR
SR
424
425
SR, single room
426
427
428
429
430
431
432
433
434
435
436
437
22
438
Figure 1b, Schematic layout of environmental sampling sites in typical high-dependency
439
bays (ward bay A, 4-bed, ward bay B, 6-bed)
440
1
1
2/6
2/6
3
4
3
5
5
3
4
4
3
5
5
4
2/6
2/6
2/6
2/6
3
4
3
5
5
3
4
4
3
5
5
2/6
2/6
2/6
3
4
3
5
5
4
2/6
6-bedded bay (B)
Nurses
Station
4-bedded bay (A)
441
1,
2,
3,
4,
5,
6,
4
Air sampling
Air settle-plate
Mattress
Pillow
Bedrail/bedframe
Locker
442
443
444
445
446
447
23
448
Table 1. Isolation of MRSA from patients and air on 132 sampling occasions
Sampling
occasions
Patients
Patient MRSA +
Air MRSA+ a
Patient MRSA + Air
MRSA - b
Air + Patient - c
Air + d
n
n
n
n
n
n
Ward bay A
29
8/116
2/29
4/29
2/29
4/29
Ward bay B
29
9/174
3/29
4/29
3/29
6/29
Ward bay C
22
3/132
1/22
2/22
1/22
2/22
Ward bay D
29
6/174
1/29
4/29
4/29
5/29
Ward bay E
16
3/96
2/16
1/16
0/16
2/16
Ward bay F
7
1/14
0/7
1/7
0/7
0/7
132
30/706 (4.2%)
9/132 (6.8%)
16/132 (12.1%)
10/132 (7.6%)
19/132(14.4%)
Ward bay
All wards
449
450
451
452
453
454
455
456
457
a
Patient MRSA + Air MRSA +, among patients who were MRSA+, the number of air sampling
MRSA+
b
Patient MRSA + Air MRSA -, among patients who were MRSA+, the number of air sampling
MRSA-,
c
Air MRSA+ and Patient -, among air sampling that was MRSA+, the number of patients
MRSAd
Air +, number of air MRSA-positive in each ward bay
458
459
460
461
462
463
464
465
466
467
468
469
24
470
Table 2. The distribution of MRSA recovered from patients, air, settle plates and environmental
471
surfaces on 32 occasions of high-dependency bays with spa typing of 42/52 isolates and DNA
472
microarray profiling of 40/52 isolates
Date
Ward a
Ward
Bay
Bay A
Bay B
Patients
(n/total)
0/4
0/6
Air
sampling
(n/total)
0/1
0/1
Air settle
plates
(n/total)
1/4
0/6
Surface
environmentalb
(n/total)
0/16
2/24
08/06/10
08/06/10
1
1
11/06/10
2
Bay A
3/6
1/1
3/6
10/24
11/06/10
14/06/10
14/06/10
16/06/10
16/06/10
23/06/10
2
3
3
4
4
5
Bay B
Bay A
Bay B
Bay A
Bay B
Bay A
0/6
1/4 d
1/5
0/4
0/6
0/4
1/1
0/1
0/1
0/1
0/1
0/1
1/6c
0/4
0/5
0/4
0/6
0/4
1/24
0/16
0/20
0/16
0/24
1/16
23/06/10
28/06/10
28/06/10
20/09/10
20/09/10
27/09/10
27/09/10
04/10/10
04/10/10
18/10/10
18/10/10
26/10/10
5
6
6
1
1
1
1
1
1
5
5
5
Bay B
Bay A
Bay B
Bay A
Bay B
Bay A
Bay B
Bay A
Bay B
Bay A
Bay B
Bay A
0/6
1/4
0/6
0/4
0/6
0/4
0/6
0/4
0/6
0/4
0/6
1/4
1/6
0/1
0/1
0/1
0/1
0/1
0/1
0/1
0/1
0/1
0/1
1/1
1/6
0/4
0/6
0/4
0/6
0/4
0/6
0/4
0/6
0/4
0/6
0/4
2/24 c
0/16
0/24
0/16
0/24
0/16
0/24
0/16
0/24
0/16
0/24
0/16
26/10/10
08/11/10
08/11/10
04/01/11
04/01/11
10/01/11
10/01/11
31/01/11
31/01/11
21/02/11
5
5
5
7
7
7
7
8
8
8
Bay B
Bay A
Bay B
Bay A
Bay B
Bay A
Bay B
Bay A
Bay B
Bay A
2/6 e
0/4
0/6
0/4
1/6
0/4
0/6
0/4
1/6
0/4
1/1
0/1
0/1
1/1
1/1
0/1
1/1
0/1
0/1
0/1
0/6
0/4
0/6
1/4
5/6
0/4
0/6
0/4
0/6
0/4
0/24
0/16
0/24
0/16
5/24
0/16
0/24
0/16
0/24
0/16
25
21/02/11
Total
8
8
wards
Bay B
32
Bays
1/6
0/1
0/6
12/161(7.5%) 8/32(25%) 12/161(7.5%)
0/24
21/644(3.3%)
473
a
Wards 1-8
474
b
Mattress, pillow, bedrail/frame or locker.
475
c
One isolate not spa typed.
476
d
Using DNA microarray, one t032 spa type isolate lacked the sec & sel gene cluster
477
e
Total 2/6 isolates MRSA-positive, one t515 spa type, one t022
478
479
480
Cell shading indicates the combined spa types and DNA microarray profiles of isolates within
the four clusters;
spa type t557; DNA microarray profile blaZ, erm(C), sec & sel, egc and IEC B
spa type t515, DNA microarray profile blaZ, erm (C), egc and IEC B
spa type t032, DNA microarray profile blaZ, erm (C), sec & sel, egc and IEC B
481
482
483
484
485
26