Download Effect of scrubs, gowns, masks on S aureus transmission ICHE Oct

infection control and hospital epidemiology
october 2007, vol. 28, no. 10
original article
Preventing the Airborne Spread of Staphylococcus aureus
by Persons With the Common Cold: Effect of
Surgical Scrubs, Gowns, and Masks
Werner E. Bischoff, MD, PhD; Brian K. Tucker, BS; Michelle L. Wallis, BS; Beth A. Reboussin, PhD;
Michael A. Pfaller, MD; Frederick G. Hayden, MD; Robert J. Sherertz, MD
objective. Transmission of Staphylococcus aureus via air may play an important role in healthcare settings. This study investigates the
impact of barrier precautions on the spread of airborne S. aureus by volunteers with experimentally induced rhinovirus infection (ie, the
common cold).
design.
setting.
Prospective nonrandomized study.
Wake Forest University School of Medicine (Winston-Salem, NC).
participants.
A convenience sample of 10 individuals with nasal S. aureus carriage selected from 593 students screened for carriage.
intervention. Airborne S. aureus dispersal was studied in the 10 participants under the following clothing conditions: street clothes,
surgical scrubs, surgical scrubs and a gown, and the latter plus a face mask. After a 4-day baseline period, volunteers were exposed to a
rhinovirus, and their clinical course was followed for 12 days. Daily swabs of nasal specimens, pharynx specimens, and skin specimens
were obtained for quantitative culture, and cold symptoms were documented. Data were analyzed by random-effects negative binomial
models.
results. All participants developed a common cold. Incidence rate ratios (IRRs) indicated that, compared with airborne levels of S.
aureus during sessions in which street clothes were worn, airborne levels decreased by 75% when surgical scrubs were worn (P ! .001), by
80% when scrubs and a surgical gown were worn (P ! .001), and by 82% when scrubs, a gown, and a face mask were worn (P ! .001).
The addition of a mask to the surgical scrubs and gown did not reduce the airborne dispersal significantly (IRR, 0.92; P 1 .05). Male
volunteers shed twice as much S. aureus as females (incidence rate ratio, 2.04; P p .013 ). The cold did not alter the efficacy of the barrier
precautions.
conclusions. Scrubs reduced the spread of airborne S. aureus, independent of the presence of a rhinovirus-induced cold. Airborne
dispersal of S. aureus during sessions in which participants wore surgical scrubs was not significantly different from that during sessions
in which gowns and gowns plus masks were also worn.
Infect Control Hosp Epidemiol 2007; 28:1148-1154
The use of special clothes in designated areas of the hospital
was initiated in the late 1900s.1 Around 1880, Sir William
MacEwen, a British surgeon and colleague of Joseph Lister,
first replaced the frock coat usually worn during surgery with
a sterilizable white apron. The rationale for doing so was to
prevent transmission of bacteria in the hospital setting and
thereby reduce the number of infections, particularly in surgical wounds. The use of sterile surgical clothing was refined
during subsequent decades and is now widely accepted in
today’s healthcare environment.2
On the basis of our previous studies regarding the increased
spread of Staphylococcus aureus by persons with a rhinovirus-
induced common cold (ie, the “cloud phenomenon”), we
became interested in the efficacy of the barrier precautions
currently used in the healthcare setting for preventing airborne dispersal, as well as in how a common cold might
impact the efficacy of these precautions.3,4
methods
Volunteers
Five hundred ninety-three volunteers from Wake Forest University (Winston-Salem, NC) were screened for nasal S. aureus
carriage between October 2000 and April 2004, with cultures
From the Sections on Infectious Diseases (W.E.B., B.K.T., M.L.W., R.J.S.) and Biostatistics (B.A.R.), Wake Forest University School of Medicine, WinstonSalem, North Carolina; the Medical Microbiology Division, University of Iowa Hospitals and Clinics, Iowa City, Iowa (M.A.P.); and the Division of Infectious
Diseases and International Health, University of Virginia Health Sciences Center, Charlottesville, Virginia (F.G.H.).
Received January 30, 2007; accepted May 14, 2007; electronically published August 29, 2007.
䉷 2007 by The Society for Healthcare Epidemiology of America. All rights reserved. 0899-823X/2007/2810-0004$15.00. DOI: 10.1086/520734
barrier precautions and airborne spread of s. aureus
of nasal specimens. A total of 120 (20%) had specimens that
yielded S. aureus in the initial culture and in the 2 subsequent
follow-up cultures performed 2 weeks apart. A self-administered questionnaire about medical and medication history
was completed by all participants. A convenience sample of
10 healthy individuals with nasal carriage of S. aureus subsequently agreed to participate in the study.
Microbiologic Analysis
Daily swabs of nasal specimens, pharynx specimens, and skin
specimens (from the cheeks, axillae, palms of the hands, and
groin) were obtained for quantitative culture. After surface
plating and incubation at 35⬚C for 24-48 hours, the total
number of colonies and the number with a morphology consistent with that of S. aureus or coagulase-negative staphylococci (CoNS) were counted; data are presented as colonyforming units (cfu). Select colonies were evaluated by Gram
stain, a catalase test, and a coagulase test (Staphaurex; Murex
Biotech).
Air samples for culture were obtained in an airtight chamber with a volume of approximately 3.1 m3 (110 ft3). The
chamber was built around the front of a class II biologic
safety hood (Purifier; Labconco) and provided sufficient
room for a volunteer to sit in front of the workbench. Three
2-stage air samplers and one 6-stage air sampler (Andersen
Instruments) were used to measure the number of airborne
infectious bacteria, by impact on blood agar plates.5 The agar
plates were incubated for 24 hours at 35⬚C and stored for an
additional 24 hours at room temperature.6 Colonies were
counted and evaluated as described above.
One colony of S. aureus from a culture of each site sampled,
if available, was stored in 95% tryptic soy broth and 5%
glycerol at ⫺70⬚C for molecular typing. Molecular typing of
selected isolates of S. aureus was performed by pulsed-field
gel electrophoresis (PFGE) (BioRad) after digestion with the
restriction endonuclease SmaI.7 Cluster analysis of banding
patterns was used to determine clonal similarity (BioNumerics software; Applied Maths). Strains were considered
to be unique if differences in more than 3 bands were detected.8 Isolates from dominant strains were tested for methicillin susceptibility by the disk diffusion test, using 1-mg
methicillin disks (control S. aureus ATCC 25923).9 Detailed
descriptions of the methods for specimen collection, virus
challenge, and assessment of rhinovirus infection and illness
are published elsewhere.3,4
Design
Air samples were collected on agar plates daily for 16 consecutive days. Each day, participants provided samples while
sitting in the chamber during four 20-minute sessions, the
first of which was preceded by a control run without the
participant. To reduce the potential carryover of airborne
bacteria from consecutive sessions, the biologic safety hood
was run for at least 10 minutes between test sessions, and
the order of the individual sessions was randomized. Partic-
1149
ipants wore their street clothes during the first session, surgical scrubs (product numbers 720PTJL-CM [shirt] and
700PTJL-CM [pants]; Medline Industries) during the second,
surgical scrubs plus a fluid-resistant gown (product number
MDT01209XL; Medline Industries) during the third, and the
latter plus a face mask (N95 Particulate Respirator; 3M) during the fourth. Surgical gloves, shoe covers, and a bouffant
cap were also worn during the final 3 sessions. Surgical scrubs
and gowns were replaced with sterile items after each session.
The participants were trained in the appropriate use of the
face masks. During chamber sessions, study personnel directly
observed and recorded the number of sneezes, coughs, nose
blows, talking events, and unusual activities of the participants. Participants were asked to avoid brisk body movements
and to sneeze or cough, if necessary, in the direction of the
safety hood without covering the face area. After the daily
test sessions the chamber was cleaned and disinfected.
After day 4, participants were exposed to 1 of 3 rhinovirus
serotypes (Hanks, 39, or 16) at an intranasal dose of approximately 100 median tissue culture infective doses, based
on susceptibility (defined as a serum-neutralizing antibody
titer of no more than 1 : 4 against the challenge virus).10,11
Infection was determined by recovery of virus from nasal
lavage specimens and by detection of positive homotypic serum-neutralizing antibody titers in paired acute-phase specimens (obtained at the time clinical symptoms presented)
and convalescent-phase specimens (obtained 4 weeks later).
To assess illness, 8 symptoms (sneezing, rhinorrhea, nasal obstruction, sore or scratchy throat, cough, malaise, chills, and
headache) were each evaluated daily, using a scale of 0-4.12
The total cold-symptom score is the sum of the individual
symptom ratings. A cold was defined as present if a subject
had a total symptom score of 6 and rhinorrhea on 3 or more
days and/or had the subjective impression of a cold.12
Informed consent was obtained from all participants. The
human experimentation guidelines of the US Department of
Health and Human Services and those of the Wake Forest
University School of Medicine were followed in the conduct
of this study.
Data Analysis
Data analyses were performed using Stata, version 9.1 (StataCorp). The primary outcome was defined as the dispersal
of airborne bacteria, such as S. aureus, CoNS, a-hemolytic
streptococci, and other bacteria, at the individual level in cfu/
m3/min. The exposure variable—clothing condition—was
transformed into indicator variables to allow multiple comparisons between the types of clothes worn. Model building
and testing was based on the random-effects negative binomial distribution for overdispersed count data.13,14 The effect
measures are expressed as incidence rate ratios (IRRs) in cfu/
m3/min. This distribution was chosen to account for the between-individual heterogeneity in dispersal rates, which can
cause overdispersion; the excess number of zero values found
in the outcome distributions; and the additional level of clus-
1150
infection control and hospital epidemiology
october 2007, vol. 28, no. 10
culation of the final models, model diagnostic analyses were
used to search for potential influential cases. Because of the
cluster nature of the data set, participants were removed one
at a time from the final models, and the change-in-estimate
effects were determined. The threshold was preset at a 20%
change-in-estimate effect.
results
Participants
figure.
Cold-symptom scores for 10 volunteers exposed to rhinovirus on day 5 of the 16-day study. Scores were calculated using
the methods of Harris and Gwaltney.12
tering present because of repeated observations occurring for
an individual over time and across clothing conditions.
Model building was based on a hierarchical backward elimination strategy, which pared down the most-complex models
to a final model. Effect measure modification was assessed
by the composite Wald test (a level, .05) as well as the Akaike
information criteria and the Schwartz information criteria.15-17 The change-in-estimate technique was chosen to test
for confounding, with a preset threshold of 10%. After cal-
Five healthy female volunteers and 5 healthy male volunteers
were enrolled, all of whom were white. The mean age was
25.4 years (range, 21-32 years). None had chronic medical
conditions or requirements associated with an increased risk
of S. aureus carriage, such as diabetes mellitus, hemodialysis,
atopic dermatitis, psoriasis, and eczema. Four volunteers reported a history of seasonal allergies but were not symptomatic and were not taking medication at the time of the chamber sessions. None had recently received or were taking
antibiotics, nasal sprays, inhaled steroids, antihistamines, and/
or corticosteroids. Two volunteers were taking hormonal
birth control agents. One volunteer was a smoker.
Rhinovirus Challenge
All participants developed a symptomatic cold after the rhinovirus challenge, according to their total cold-symptom
table 1. Findings of Random-Effects Negative Binomial Models to Compare Bacterial Carriage
at 4 Body Sites Before and After Rhinovirus Inoculation
No. of cfu, mean Ⳳ SD
Site, bacteria type
Nose
S. aureus
CoNS
a-HS
Other
All
Pharynx
S. aureus
CoNS
a-HS
Other
All
Cheeks
S. aureus
CoNS
a-HS
Other
All
Skin
S. aureus
CoNS
a-HS
Other
All
note.
Before
Incidence rate ratio
(95% CI)
After
P
5,308.38
3,682.34
32.73
2,477.12
2,803.59
Ⳳ
Ⳳ
Ⳳ
Ⳳ
Ⳳ
1,443.92
552.90
23.51
601.29
501.01
3,558.84
3,173.61
441.62
4,516.59
2,793.38
Ⳳ
Ⳳ
Ⳳ
Ⳳ
Ⳳ
738.96
274.14
142.21
730.93
275.98
0.61 (0.47-0.80)
NA
3.38 (1.00-11.45)
1.90 (1.26-2.88)
1.17 (0.96-1.44)
6.23
114.36
7,728.73
7,114.20
4,272.13
Ⳳ
Ⳳ
Ⳳ
Ⳳ
Ⳳ
2.91
42.57
1,479.00
1,350.76
688.53
20.83
39.89
12,357.01
9,778.99
5,545.81
Ⳳ
Ⳳ
Ⳳ
Ⳳ
Ⳳ
9.34
10.29
1,778.01
1,370.00
691.49
1.22
0.85
1.18
1.06
1.00
(0.72-2.08)
(0.53-1.38)
(0.90-1.55)
(0.81-1.40)
(0.82-1.20)
.468
.517
.229
.662
.922
14.47
1,104.92
62.64
298.02
363.71
Ⳳ
Ⳳ
Ⳳ
Ⳳ
Ⳳ
2.82
132.48
25.23
63.79
43.49
0.82
1.04
1.31
1.13
1.05
(0.55-1.22)
(0.82-1.31)
(0.80-2.15)
(0.85-1.52)
(0.87-1.26)
.328
.750
.284
.407
.624
15.18
265.80
76.10
370.52
196.70
Ⳳ
Ⳳ
Ⳳ
Ⳳ
Ⳳ
4.45
73.19
26.89
85.48
35.68
0.75
1.06
1.56
1.41
1.10
(0.47-1.88)
(0.80-1.40)
(0.93-2.62)
(1.08-1.85)
(0.91-1.33)
.221
.704
.092
.013
.310
11.23 Ⳳ 3.24
901.09 Ⳳ 152.35
20.69 Ⳳ 8.22
260.7 Ⳳ 66.39
302.50 Ⳳ 54.35
10.38
370.15
2.89
188.52
90.36
Ⳳ
Ⳳ
Ⳳ
Ⳳ
Ⳳ
3.96
232.45
0.87
39.81
14.82
!.001
.050
.002
.117
a-HS, a-hemolytic streptococci; CI, confidence interval; cfu, colony-forming units; CoNS, coagulasenegative staphylococci; NA, convergence was not achieved; S. aureus, Staphylococcus aureus.
barrier precautions and airborne spread of s. aureus
table 2.
Bacteria
type
S. aureus
CoNS
a-HS
Other
All
1151
Airborne Dispersal of Bacteria Before and After Rhinovirus Inoculation, by Clothing Condition
Surgical scrubs,
cfu/m3/min,
mean Ⳳ SD
Street clothes,
cfu/m3/min,
mean Ⳳ SD
Before
0.95
27.34
6.73
107.89
142.90
Ⳳ
Ⳳ
Ⳳ
Ⳳ
Ⳳ
2.25
37.74
33.35
335.95
380.21
After
0.73
27.05
3.99
87.54
119.31
Ⳳ
Ⳳ
Ⳳ
Ⳳ
Ⳳ
1.39
50.05
13.73
249.08
263.12
Before
0.17
7.70
0.27
7.36
15.49
Ⳳ
Ⳳ
Ⳳ
Ⳳ
Ⳳ
0.36
9.00
0.71
7.69
15.67
After
0.15
8.64
8.13
24.67
41.58
Ⳳ
Ⳳ
Ⳳ
Ⳳ
Ⳳ
0.38
18.19
74.51
100.22
169.24
Surgical scrubs and gown,
cfu/m3/min,
mean Ⳳ SD
Surgical scrubs, gown,
and mask,
cfu/m3/min,
mean Ⳳ SD
Before
Before
0.13
2.88
0.30
4.08
7.39
Ⳳ
Ⳳ
Ⳳ
Ⳳ
Ⳳ
0.27
3.60
0.90
4.16
7.55
After
0.10
3.98
12.23
44.37
60.68
Ⳳ
Ⳳ
Ⳳ
Ⳳ
Ⳳ
0.29
8.68
116.40
306.69
369.90
0.30
5.30
0.07
5.43
11.09
Ⳳ
Ⳳ
Ⳳ
Ⳳ
Ⳳ
0.99
7.34
0.19
5.45
12.17
After
0.07
2.95
0.13
4.21
7.35
Ⳳ
Ⳳ
Ⳳ
Ⳳ
Ⳳ
0.19
4.32
0.33
8.40
11.22
note. No bacteria were detected during control sessions performed for each patient daily during the study. No significant differences in dispersal before
and after viral inoculation were detected for any of the clothing conditions (P 1 .05 , by random-effects negative binomial modeling). a-HS, a-hemolytic
streptococci; cfu, colony-forming units; CoNS, coagulase-negative staphylococci; S. aureus, Staphylococcus aureus.
scores. The Figure documents daily mean symptom scores
for all participants after rhinovirus exposure. Volunteers most
often experienced nasal congestion, sore throat, and runny
nose, followed by malaise, coughing, and headache. Sneezing
and chills were reported least frequently.
For 2 participants, nasal lavage specimens were not tested
and serum titers not measured because of technical problems.
Four of the remaining 8 participants had serum antibody
titers greater than 4 after viral challenge, and all 8 had rhinoviruses detected in at least 1 nasal lavage specimen. Sneezes
and coughs were rarely documented. A total of 23 sneezes
and 186 coughs during 640 chamber sessions, with no statistically significant difference before and after rhinovirus exposure. Reading was the only “unusual activity” observed by
the study personnel.
Bacteria in Mucosal and Skin Specimens
All participants carried S. aureus in their nose before and
after rhinovirus challenge. Table 1 shows the findings of statistical analysis of bacterial carriage before and after rhinovirus inoculation. Significant differences in the results of
quantitative cultures of nasal samples were observed after the
viral challenge, with a 39% decrease in the number of S.
aureus colony-forming units, a greater than 3-fold increase
in the number of a–hemolytic streptococcal colony-forming
units, and a 2-fold increase in the number of colony-forming
units of other bacteria. In addition, the mean number of
colony-forming units of other bacteria increased by 41% after
the rhinovirus challenge in cultures of skin samples obtained
from the participants.
Effects of Clothing Conditions
During 640 chamber sessions, a total of 32,946 airborne cfu
were detected, including 140 cfu of S. aureus (0.42%), 5,677
cfu of CoNS (17.23%), 5,625 cfu of a-hemolytic streptococci
(17.07%), and 21,504 cfu of other bacteria (65.27%). Samples
from all control runs were negative for airborne bacteria.
Table 2 lists the mean dispersal values for bacteria, stratified
by clothing condition. The maximum airborne dispersal varied from 1.32 to 5.74 cfu/m3/min for S. aureus, from 3.09 to
154.06 cfu/m3/min for CoNS, from 1.32 to 1,268.24 cfu/m3/
min for a-hemolytic streptococci, and from 5.30 to 3,083.85
cfu/m3/min for other bacteria. The lowest amounts of bacteria
were collected during the sessions in which surgical scrubs,
a gown, and a mask were worn, with a mean dispersal of
0.12 cfu/m3/min for S. aureus, 3.54 cfu/m3/min for CoNS,
0.11 cfu/m3/min for a-hemolytic streptococci, and 4.51 cfu/
m3/min for other bacteria. The diameter of the majority
(77%) of air particles carrying bacteria was greater than 5
mm, independent of clothing conditions and rhinovirus
exposure.
Table 3 lists findings of the final models to compare the
effect of clothing conditions and sex on airborne dispersal of
bacteria; models were run after testing for effect measure
modifiers and confounders. These included modifiable risk
factors, such as allergies and rhinovirus-induced common
cold, and fixed risk factors, such as sex and age.
Compared with airborne levels of S. aureus during sessions
in which street clothes were worn, airborne levels decreased
by 75% when surgical scrubs were worn (P ! .001), by 80%
when scrubs and a surgical gown were worn (P ! .001), and
by 82% when scrubs, a gown, and a face mask were worn
(P ! .001). However, the addition of a mask to the surgical
scrubs and gown did not reduce the airborne dispersal significantly (IRR, 0.92; P p .759). Similar reduction patterns
were also found for all other bacteria except a-hemolytic
streptococci. Here, use of a face mask led to a statistically
significant further reduction of 65% in airborne spread (IRR,
0.35; P ! .001). No effect-measure modifier was identified for
any of the outcomes. The only confounding factor detected
was sex. S. aureus was shed into the air twice as frequently
by men, compared with women, during the street clothing
sessions (IRR, 2.04). None of the other bacteria groups were
affected by sex or other factors. Additional tests revealed that
data from no particular participant significantly influenced
the outcomes of analyses involving clothing conditions.
Effects of the Common Cold
The effect of the common cold on the association between
clothing conditions and airborne bacterial spread was investigated by the inclusion of 2 variables in the multivariate
models. A binomial variable indicated exposure to the rhi-
1152
infection control and hospital epidemiology
october 2007, vol. 28, no. 10
table 3. Findings of Final Random-Effects Negative Binomial Models to Compare the Effect of Clothing Conditions and Sex on the
Airborne Dispersal of Bacteria
Staphylococcus aureus
Variable
Surgical scrubs vs
street clothes
Surgical scrubs and
gown vs street
clothes
Surgical scrubs, gown,
and mask vs street
clothes
Surgical scrubs and
gown vs surgical
scrubs
Surgical scrubs, gown,
and mask vs surgical scrubs
Surgical scrubs, gown,
and mask vs surgical scrubs and
gown
Male sex vs female sex
note.
a-hemolytic
streptococci
CoNS
Other bacteria
All bacteria
IRR (95% CI)
P
IRR (95% CI)
P
IRR (95% CI)
P
IRR (95% CI)
P
IRR (95% CI)
P
0.25 (0.17-0.37)
!.001
0.38 (0.32-0.45)
!.001
0.68 (0.50-0.94)
.019
0.34 (0.29-0.41)
!.001
0.37 (0.31-0.43)
!.001
0.20 (0.13-0.30)
!.001
0.22 (0.19-0.27)
!.001
0.45 (0.32-0.65)
!.001
0.24 (0.20-0.29)
!.001
0.25 (0.21-0.29)
!.001
0.18 (0.12-0.28)
!.001
0.23 (0.19-0.27)
!.001
0.24 (0.16-0.37)
!.001
0.23 (0.19-0.28;
!.001
0.23 (0.19-0.28)
!.001
0.78 (0.47-1.29)
.334
0.59 (0.48-0.72)
!.001
0.67 (0.46-0.97)
.032
0.70 (0.57-0.86)
.001
0.67 (0.55-0.81)
!.001
0.72 (0.43-1.20)
.205
0.60 (0.48-0.73)
!.001
0.35 (0.22-0.55)
!.001
0.67 (0.54-0.82)
!.001
0.62 (0.51-0.76)
!.001
0.92 (0.54-1.58)
2.04 (1.16-3.57)
.759
.013
1.01 (0.81-1.27)
…
.912
0.53 (0.33-0.84)
…
.008
0.96 (0.77-1.19)
…
.693
0.94 (0.76-1.14)
.513
…
CI, confidence interval; CoNS, coagulase-negative staphylococci; IRR, incidence rate ratio.
novirus at day 4, and the subjective daily symptom scores
provided a continuous measure of cold symptoms. A test of
interaction terms with these factors and change-in-estimate
assessments for confounding did not reveal a significant alteration of the exposure-outcome relationship shown above.
This held true for all bacteria groups and species.
S. aureus Characteristics
Six hundred eighty-four (8.5%) of 808,505 S. aureus colonyforming units detected on culture were studied by PFGE. Each
participant carried 1 dominant clone of S. aureus (genetic
similarity, 73.6%-100%) recovered from different sites. All S.
aureus strains were susceptible to antibiotics.
discussion
S. aureus is currently one of the most important healthcareassociated pathogens in the United States.18 This fact is further
complicated by the emergence of multidrug resistance and
the increasing frequency of community-acquired S. aureus
infection.19
Several guidelines have been published on the management
of S. aureus carriers in healthcare settings.20,21 Besides physical
isolation, cohorting, and the use of eradication treatment,
one of the main strategies is to control bacterial spread by
using barrier precautions, such as special clothing and/or face
masks.22 Transmission of S. aureus is thought to occur predominantly by direct contact through hands and, to a lesser
degree, by indirect contact via environmental surfaces or by
airborne dispersal.23-25 The latter—the target of our study—
has been studied since the middle of the last century.26,27 As
an early example, Bethune et al.28 found that the quantity of
bacteria dispersed in the air by volunteers wearing street
clothes was reduced by 18.9% after volunteers donned paper
covers over their street clothes. This and similar results from
subsequent studies confirm the findings in our study. The
highest percentage decrease in airborne dispersal was observed after volunteers changed from “contaminated” street
clothes to sterilized ”clean” clothes, underscoring the major
importance of clothes in the spread of bacteria via air. A
comparison of the reduction in rates after changing into sterilized surgical scrubs, according to bacteria type, revealed that
the reduction was most prominent for S. aureus. The addition
of a surgical gown did not alter the outcome for S. aureus
significantly but did so for the other bacteria. Apparently, S.
aureus dispersal is strongly associated with an initial change
to sterilized scrubs, whereas the spread of other bacteria can
be further reduced by additional skin coverage, particularly
of the forearms, during gown use. This hypothesis is supported by the overall lower carriage rates of S. aureus on the
skin, compared with rates for other bacteria, such as CoNS,
in our study (Table 1) and by data presented by Williams29
and Wertheim et al.30 In regard to the management of S.
aureus carriers, the use of scrubs appears to reduce the quantity of S. aureus dispersed through the air, whereas the addition of gowns does not appear to improve efficacy of scrubs.
In contrast to the efficacy of scrubs and gowns, there is
only weak evidence of the efficacy of face masks.31,32 The most
comprehensive study was undertaken by Tunevall et al.,33 who
used a prospective randomized study design to investigate the
effect of masks on surgical site infection rates among 3,088
barrier precautions and airborne spread of s. aureus
patients. Their finding, which was surprising at the time, was
that masks did not alter infections rates. Our study showed
a significant decrease in the airborne spread of a-hemolytic
streptococci during sessions in which face masks were used.
However, the spread of other bacteria was not significantly
affected by mask use. The latter finding was also confirmed
by our previous investigations, which comprised different volunteers, involved a mask with a lower filter efficiency (75%
filtration capacity), and did not include a session during
which surgical scrubs and a gown were worn.3,4 A probable
explanation is that a-hemolytic streptococci colonized only
the pharynx and mouth and were therefore blocked from
dispersing through the air by N95 face masks, in contrast to
other types of bacteria, which were also found on and spread
from the skin and/or clothes of the participants (Table 1).
This implies that, for S. aureus, the nose might play a role
as a source for contamination of the skin and/or clothes or
for transmission by indirect contact via hands but not for
direct contact via the airborne route.
Another interesting finding is the influence of sex. Male
volunteers spread twice the amount of S. aureus via the air,
compared with female volunteers, independent of clothing
conditions. Earlier studies confirm this sex-based association,
showing that 12% of male volunteers disperse S. aureus, compared with only 1% of female volunteers.28,34 However, the
mechanisms behind these findings remain unclear.
The second objective of our study was to determine the
impact of a rhinovirus-induced common cold on the efficacy
of barrier precautions. The motivation for evaluating this
objective was based on the finding of increased S. aureus
dispersal by persons with a viral upper respiratory tract
infection.3,4,10,35 In contrast to our earlier findings,3 this study
revealed that, for all protective clothing conditions tested, the
airborne spread of S. aureus after the virus challenge was less
than that before the virus challenge. We found that a mildto-moderate cold did not alter the efficacy of the barrier
precautions.
One must be aware of several constraints regarding the
interpretation of the results. First, no clear threshold currently
exists for the airborne infective dose of S. aureus. Therefore,
the clinical significance of our findings cannot be directly
assessed at this time. Second, the number of participants was
restricted to 10 young and healthy volunteers and to certain
brands and types of barrier precautions. Similar restrictions
are not uncommon in viral challenge trials, and they may
somewhat limit the generalizability of our findings. Third,
volunteers rated the symptoms of the rhinovirus-induced cold
as mild to moderate. The impact of a severe viral upper
respiratory tract infection may therefore differ from that reported by our participants. Fourth, only S. aureus carriers
were enrolled into the study, which makes inferences about
other potentially airborne pathogens, such as pneumococci,
Neisseria meningitidis, and group A streptococci, difficult to
formulate.36-39 Finally, the effects of wearing clothes over time
could not be addressed. Charnley and Eftekhar40 showed that
1153
skin bacteria penetrated through clothes worn for extended
periods. We believe that surgical scrubs, at least, would show
similar penetration over time, although it is unclear how long
this might take to occur.
This interventional study targeted the effect of commonly
used barrier precautions on the airborne dispersal of S. aureus
by persons with a rhinovirus-induced common cold. The
most efficient reduction in the airborne spread of S. aureus
was obtained by changing into sterilized surgical scrubs. The
efficacy observed during the surgical scrubs sessions was not
significantly different from that during the surgical scrubs
and gown sessions and the surgical scrubs, gown, and N95
mask sessions. However, even under the most effective clothing conditions (surgical scrubs, gown, and mask), a mean of
0.12 cfu/m3/min of S. aureus were spread into the environment. Future guidelines for nonsurgical settings and surgical
settings should take into account the importance of wearing
clean clothes during contact with S. aureus carriers, including
carriers with multidrug-resistant strains. Furthermore, mildto-moderate common colds do not impact the efficacy of
commonly used barrier precautions. Special accommodations, such as mandatory face mask use by S. aureus carriers,
appear unjustified in view of findings on the airborne dispersal of this pathogen.
acknowledgments
We thank D. Thacker (Division of Epidemiology and Virology, University
of Virginia Health Sciences Center, Charlottesville) for professional work
with the rhinovirus specimens, and L. Boyken and R. Hollis (University of
Iowa, Iowa City) for performing molecular typing analyses on the bacterial
isolates. We are very grateful for their dedication to this project.
Financial support. This study was funded by the National Institutes of
Health (grant 1R01AI46558-01A1 [ie, the “Cloud Adult” study]).
Potential conflicts of interest. All authors report no conflicts of interest
relevant to this article.
Address reprint requests to Werner E. Bischoff, MD, PhD, Wake Forest
University School of Medicine, Department of Internal Medicine, Section
on Infectious Diseases, Medical Center Boulevard, Winston-Salem, NC
27157-1042 ([email protected]).
Presented in part: 46th Interscience Conference on Antimicrobial Agents
and Chemotherapy; San Francisco, CA; September 27– September 30, 2006
(Abstract K-1676).
references
1. Fox NJ. Scientific theory choice and social structure: the case of Joseph
Lister’s antisepsis, humoral theory and asepsis. Hist Sci 1988; 26:367-397.
2. Whyte W, Vesley D, Hodgson R. Bacterial dispersal in relation to operating room clothing. J Hyg (Lond) 1976; 76:367-378.
3. Bassetti S, Bischoff WE, Walter M, et al. Dispersal of Staphylococcus aureus
into the air associated with a rhinovirus infection. Infect Control Hosp
Epidemiol 2005; 26:196-203.
4. Bischoff WE, Bassetti S, Bassetti-Wyss BA, et al. Airborne dispersal as a
novel transmission route of coagulase-negative staphylococci: interaction
between coagulase-negative staphylococci and rhinovirus infection. Infect
Control Hosp Epidemiol 2004; 25:504-511.
1154
infection control and hospital epidemiology
october 2007, vol. 28, no. 10
5. Andersen AA. New sampler for the collection, sizing, and enumeration
of viable airborne particles. J Bacteriol 1958; 76:471-484.
6. White A, Hemmerly T, Martin MP, et al. Studies on the origin of drugresistant staphylococci in a mental hospital. Am J Med 1959; 27:26-39.
7. Hollis RJ, Bruce JL, Fritschel SJ, et al. Comparative evaluation of an
automated ribotyping instrument versus pulsed-field gel electrophoresis
for epidemiological investigation of clinical isolates of bacteria. Diagn
Microbiol Infect Dis 1999; 34:263-268.
8. Tenover FC, Arbeit RD, Goering RV, et al. Interpreting chromosomal
DNA restriction patterns produced by pulsed-field gel electrophoresis:
criteria for bacterial strain typing. J Clin Microbiol 1995; 33:2233-2239.
9. National Committee for Clinical Laboratory Standards (NCCLS). Performance Standards for Antimicrobial Susceptibility Testing: 12th Informational Supplement. Wayne, PA: NCCLS; 2002:M100-S12.
10. Sherertz RJ, Reagan DR, Hampton KD, et al. A cloud adult: the Staphylococcus aureus–virus interaction revisited. Ann Intern Med 1996; 124:
539-547.
11. Stone AA, Bovbjerg DH, Neale JM, et al. Development of common cold
symptoms following experimental rhinovirus infection is related to prior
stressful life events. Behav Med 1992; 18:115-120.
12. Harris JM II, Gwaltney JM Jr. Incubation periods of experimental rhinovirus infection and illness. Clin Infect Dis 1996; 23:1287-1290.
13. Long JS. Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: Sage; 1997.
14. Cameron AC, Trivedi PK. Regression Analysis of Count Data. Cambridge:
Cambridge University Press; 1998.
15. Akaike H. Information theory and an extension of the maximum likelihood principle. In: Petrov BN, Csaki F, eds. Second International Symposium on Information Theory. Budapest: Akademiai Kaido; 1973:267281. Reprinted in: Kotz S, Johnson NL, eds. Breakthroughs in Statistics.
Vol. 1. New York: Springer; 1992:599-624.
16. Schwarz G. Estimating the dimension of a model. Ann Stat 1978; 6:461464.
17. Burnham KP, Anderson DR. Model Selection and Inference: A Practical
Information-Theoretic Approach. New York: Springer; 1998.
18. National Nosocomial Infections Surveillance (NNIS) report, data summary from October 1986-April 1996, issued May 1996: a report from
the National Nosocomial Infections Surveillance (NNIS) System. Am J
Infect Control 1996; 24:380-388.
19. Fridkin SK, Hageman JC, Morrison M, et al. Methicillin-resistant Staphylococcus aureus disease in three communities. Active Bacterial Core Surveillance Program of the Emerging Infections Program Network. N Engl
J Med 2005; 352:1436-1444.
20. Muto CA, Jernigan JA, Ostrowsky BE, et al. SHEA guideline for preventing nosocomial transmission of multidrug-resistant strains of Staphylococcus aureus and enterococcus. Society for Healthcare Epidemiology
of America. Infect Control Hosp Epidemiol 2003; 24:362-386.
21. Garner JS. Guideline for isolation precautions in hospitals. The Hospital
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
Infection Control Practices Advisory Committee. Infect Control Hosp
Epidemiol 1996; 17:53-80.
Centers for Disease Control and Prevention. Interim guidelines for prevention and control of staphylococcal infection associated with reduced
susceptibility to vancomycin. MMWR Morb Mortal Wkly Rep 1997; 46:
626-628, 635-636.
John JF Jr, Barg NL. Staphylococcus aureus. In: Mayhall CG, ed. Hospital
Epidemiology and Infection Control. 3rd ed. Philadelphia: Lippincott Williams and Wilkins; 2004:443-470.
Williams RE. Epidemiology of airborne staphylococcal infection. Bacteriol
Rev 1966; 30:660-674.
Rammelkamp CH Jr, Mortimer EA Jr, Wolinsky E. Transmission of streptococcal and staphylococcal infections. Ann Intern Med 1964; 60:753-758.
Hare R, Thomas CG. The transmission of Staphylococcus aureus. Br Med
J 1956; 12:840-844.
Koch F. Perspectives on barrier material standards for operating rooms.
Am J Infect Control 2004; 32:114-116.
Bethune DW, Blowers R, Parker M, et al. Dispersal of Staphylococcus
aureus by patients and surgical staff. Lancet 1965; 1:480-483.
Williams RE. Healthy carriage of Staphylococcus aureus: its prevalence
and importance. Bacteriol Rev 1963; 27:56-71.
Wertheim HF, Melles DC, Vos MC, et al. The role of nasal carriage in
Staphylococcus aureus infections. Lancet Infect Dis 2005; 5:751-762.
Lipp A, Edwards P. Disposable surgical face masks for preventing surgical
wound infection in clean surgery. Cochrane Database Syst Rev 2002:
CD002929.
Romney MG. Surgical face masks in the operating theatre: re-examining
the evidence. J Hosp Infect 2001; 47:251-256.
Tunevall TG. Postoperative wound infections and surgical face masks: a
controlled study. World J Surg 1991; 15:383-387.
Hill J, Howell A, Blowers R. Effect of clothing on dispersal of Staphylococcus aureus by males and females. Lancet 1974; 9: 1131-1133.
Eichenwald HF, Kotsevalov O, Fasso LA. The “cloud baby”: an example
of bacterial-viral interaction. Am J Dis Child 1960; 100:161-173.
Nichol KP, Cherry JD. Bacterial-viral interrelations in respiratory infections of children. N Engl J Med 1967; 277:667-672.
Gwaltney JM, Sande MA, Austrian R, et al. Spread of Streptococcus pneumoniae in families: relation of transfer of S. pneumoniae to incidence of
colds and serum antibody. J Infect Dis 1975; 132:62-68.
Harrison LH, Armstrong CW, Jenkins SR, et al. A cluster of meningococcal disease on a school bus following epidemic influenza. Arch Intern
Med 1991; 151:1005-1009.
Gwaltney JM, Hayden FG. The nose and infection. In: Proctor DF, Andersen I, eds. The Nose: Upper Airway Physiology and the Atmospheric
Environment. Amsterdam: Elsevier Biomedical Press; 1982:399-422.
Charnley J, Eftekhar N. Penetration of gown material by organisms from
the surgeon’s body. Lancet 1969; 1:172-173.