Laboratory
Animals
http://lan.sagepub.com/
Evaluation of a one-way airflow system in an animal room based on counts of airborne
dust particles and bacteria and measurements of ammonia levels
C. Yamauchi, T. Obara, N. Fukuyama and T. Ueda
Lab Anim 1989 23: 7
DOI: 10.1258/002367789780886849
The online version of this article can be found at:
http://lan.sagepub.com/content/23/1/7
Published by:
http://www.sagepublications.com
On behalf of:
Laboratory Animals LtdLaboratory Animals Ltd
Additional services and information for Laboratory Animals can be found at:
Email Alerts: http://lan.sagepub.com/cgi/alerts
Subscriptions: http://lan.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
>> Version of Record - Jan 1, 1989
What is This?
Downloaded from lan.sagepub.com at PENNSYLVANIA STATE UNIV on March 3, 2014
7
Laboratory Animals (1989) 23, 7-15
Evaluation of a one-way airflow system in an animal room
based on counts of airborne dust particles and bacteria
and measurements of ammonia levels
C. YAMAUCHI, T. OBARA, N. FUKUYAMA & T. UEDA
Institute of Laboratory Animal Sciences, Faculty of Medicine, Kagoshima University, 1208-1,
Usuki-cho, Kagoshima 890, Japan
Summary
Air cleanliness in the working area of an animal
room equipped with a conventional turbulent
flow air distribution systems was compared with
that in a similar room fitted with a one-way-flow
air distribution system; in this, the supply air
flowed from the working area through the racks
of cages and was removed from the exhaust side.
Before the introduction of animals, the air in
the working and exhaust areas of both rooms
was ascertained to be Class 100. With animals
in situ, however, whereas in the turbulent airflow
room both the workspace and exhaust air
reached about Class 10000 (with particle counts,
bacterial counts and ammonia levels being
almost the same) in the one-way-flow room, the
air in the work space only went up to about Class
1000.
With the addition of sliding doors or curtains
in the personnel working area and within 60 min
on the exhaust side.
Keywords: Animal room; Environment; Airborne
dust particles; Airborne bacteria; Ammonia;
Rats
Allergies to laboratory animals (ALA) have
become a problem amongst personnel working
in close contact with laboratory animals and the
development of asthma as a symptom of ALA
has been recognized as a prescribed industrial
disease in some countries for several years (e.g.
United Kingdom, 1981: HMSO Command 8121).
The incidence amongst susceptible workers has
been reported to range from 11 to 32% (Lincoln
et al., 1974; Lutsky & Neuman, 1975; Gross,
1980; Cockcroft et al., 1981; Davies & McArdle,
in front of the rack in the one·way·flow room 1981; Slovak & Hill, 1981; Agrup et al., 1986;
the work space air was maintained at Class 100
Botham & Teasdale, 1987; Study Group on
Laboratory Animal Allergy, 1987) and it is clear
with almost no dust particles over 1"m in size,
airborne bacteria or ammonia being detectable.
that allergens such as fur, dandruff and urinary
A comparison of all factors measured showed
proteins are present in the dust in the air of
animal rooms. However, there have been few
that whereas in the turbulent flow room the
contamination of the work space air was 910,10 published studies on air conditioning methods
of that of the exhaust air, in the one-way-flow
aimed at reducing airborne allergens in animal
room working areas.
room it was only 47%, with curtains added this
was reduced to 70,10
and with sliding doors to only
Except in the case of hazardous work involving
2%. In the latter case, contamination levels
infections, radioactivity, airborne toxins etc.,
turbulent flow air conditioning systems have
increased markedly on both sides during and
immediately after cage changing, but recovered
generally been used in animal rooms because
they provide quite good temperature and
to the pre-cage changing levels within 30 min
humidity distributions. With such systems,
Received 3 December 1987; accepted 9 May 1988
airborne dost, bacteria and ammonia from the
Downloaded from lan.sagepub.com at PENNSYLVANIA STATE UNIV on March 3, 2014
8
Yamauchi, Obara, Fukuyama, Veda
cages are necessarily dispersed throughout the
room air.
This study reports on a one-way-flow air
conditioning system which has been developed
so that the clean supply air flows from the
working area of the animal room, through the
cage racks and out from the exhaust side; some
results have already been reported (Yamauchi et
al., 1986). The effectiveness of this system in
maintaining air cleanliness (in terms of particle
counts, microorganism counts and ammonia
levels) was assessed by comparing the levels of
these factors in the work space and exhaust air in:
(1) a turbulent airflow (TAF) room; (2) a room
equipped with the one-way-airflow (OWAF)
system with open-fronted racks; (3) a room
similar to (2) but with perforated curtains in
front of the racks; and (4) a room similar to (2)
but with perforated sliding doors in front of the
racks.
Materials and methods
Animal rooms and racks
Two experimental rooms were set up for this
study. Each room was 1590 x 1785x 2120 mm
high. One was provided with the traditional TAF
system, the other with the OWAF system.
Construction and air conditioning details have
been reported previously (Yamauchi et al., 1986).
In both systems HEPA filtered air was supplied
through slit-type diffusers in the ceiling. In the
TAF room, the wall mounted exhaust outlets
were situated 285 mm above the floor following
the conventional system in common use in
Japan. The stainless steel rack (450 x 1540x
1600 mm) in this room had 5 levels and
accommodated a total of 20 plastic cages.
Figure 1 shows, diagrammatically, the OWAF
animal room system. As already noted the
personnel working area was ventilated in the
same way as that in the TAF room. In this
OWAF room, however, the personnel area was
completely separated from the exhaust side by
valance plates fitted above, below and to the sides
of the rack in order to maximize the likelihood
of a one-way flow of air through it. Here also,
the exhaust vents were mounted on the ceiling
Fig. 1. Diagrammatic representation of the one-way-airnow
animal room with sliding doors. A, supply air inlet; B, sliding
doors; C, air control plate; D, adjustable air exhaust slit; E,
exhaust air outlet.
of the exhaust plenum created by the valance
plates around the rack. Again, the stainless steel
rack (450 x 1540x 1670mm) had 5 levels accommodating 20 plastic cages. Adjustable air exhaust
slits were provided on the back of the rack just
above each cage level. Three millimetres thick,
plastic airflow control plates (1480 x 105 mm)
were fixed to the racking above the cages as
indicated in Fig. 1 to encourage air to flow into
the cages. Results from the traditional TAF room
were compared with those from an OWAF room
fitted with either:
(1) an open fronted rack without airflow control
plates.
(2) a rack fitted with two vinyl curtains (0' 3 mm
thick, 795 x 1670 mm) perforated by 40 50-mm
diameter holes arranged in horizontal rows to coincide with the top of each row of cages; the ends
of the curtains were attached firmly to the rack by
magnets and airflow control plates were provided.
(3) A rack with 4 perforated, acrylic sliding
doors (380 x 280 mm) fitted at each level. Each
Downloaded from lan.sagepub.com at PENNSYLVANIA STATE UNIV on March 3, 2014
One-way airflow system in animal rooms
door had 4 50 mm diameter holes at cage top
level and again airflow control plates were fitted.
Animals and rearing conditions
In every case 60 9-18 week old SPF male rats
(Kud : Wister) were used. Three of these were
accommodated in each 300x 355x 175mm plastic
cage with a steel mesh cover with Japanese cedar
sawdust as bedding. Both cages and bedding
were autoclaved at 121°C for 30 min and
replaced weekly. Food (CLEA Japan, Tokyo
CE-2) and tap water were given ad libitum.
Environmental conditions
The rooms were maintained at 22 ± 1 °C and
50± 10% r.h. Air change rates were 16 times per
hour in the TAF room and 8 times per hour in
the OWAF room as calculated from the supply
air volumes. This difference was based on the
results of earlier work (Yamauchi et al., 1986)
which showed that even with the lower air change
rate, equivalent control of air temperature,
relative humidity and air velocity was achievable
in both rooms; also there was no back flow of
air from the racks in the OWAF room other than
when the rack was open to the room.
Measurement of air cleanliness
The counts ofthe numbers of 0'3,0'5, 1,0,2,0
and 5·0 ILm particles per litre of air were made
on the working area and exhaust side simultaneously using two similar particle counters and
recorders (Ryon, Tokyo, Model KM-02 and
KP-04, respectively). Measurements were carried
out every 2 min during the period 0900 to 1000h,
and the average of these counts is quoted as the
measured value. In order to determine the
classification level according to Federal Standard
209B (1973), the counts were also converted
mathematically to particle numbers/foot3•
The number of bacteria was determined using
a pinhole sampler (Sanki Science, Tokyo, Model
SY) whereby 531 of air is blown over tryptosoy
agar in 90 mm Petri dishes in a 2 min period.
After incubation at 37°C for 48 h, the number
of colony forming units (CFU) was counted and
converted to numbers/IOO I of air.
9
In order to avoid the disturbing effects which
would have been caused by operatives working
in the experimental areas, both the particle
counters and pinhole sampler were operated from
outside the rooms. Sample air (from 1m above
the centre of the floor on both the personnel and
exhaust sides of the rack) was delivered to the
equipment through vinyl tubing 10mm diameter
and 2 m in length. On completion of sampling for
airborne bacteria, settle plate samples were collected using the Koch method (Committee on
Standards of Laboratory Animal Facilities, 1982).
This involved leaving open, 90 mm Petri dishes
containing tryptosoy agar on the floor of both the
personnel and exhaust side for 30 min from 1000
to 1030h. This operation was carried out by staff
entering the room wearing dust-proof suits. After
collection the plates were incubated at 37°C for
48 h and the number of CFU's counted.
Ammonia concentration was measured at points
1m above the centre of the floor on the
personnel and exhaust sides and 50 mm above the
centre of the cage floors, using a Kitagawa-type
gas detector (Komyo Science, Kawasaki, Models
SC and SD).
The measurements of bacteria and ammonia
levels were carried out daily and in duplicate at
each sampling point both before and for 7 days
after animals were introduced; here the average
of the duplicate measurements is quoted as the
measured value. Because airborne particle and
bacterial counts in animal rooms are known to
show a circadian rhythm (Obara et al., 1986),
measurements were performed at fixed times
between 0900 and 1100 h.
More intensive studies of the numbers of
airborne particles, CFU's and ammonia levels
occurring before and after cage changing were
made only in a OW AF room using a rack fitted
with sliding doors. In this case, samples were
collected from both the personnel and exhaus.t
sides of the rack 7 days after the animals had
been introduced:
(a) before cage changing; (b) about 30 min after
cage cleaning commenced; (c) immediately after
cage cleaning was completed; and (d) at 15, 30,
45 and 60 min afterwards.
Downloaded from lan.sagepub.com at PENNSYLVANIA STATE UNIV on March 3, 2014
Yamauchi, Obara, Fukuyama, Veda
10
Statistical analysis
particles/foot3
of air from either side of each
system for the 7 days after the introduction of
animals. Air cleanliness was Class 10 000 on both
sides of the rack in the T AF room and Class 3000
and 8000 respectively on the personnel and
exhaust side of the OW AF room with the openfronted rack. In the OW AF rooms with curtains
or sliding doors, however, the air in the personnel
area remained at Class 100 throughout the 7-day
period-except
during and for 15 min after cage
changing.
For each parameter, the level measured in the air
from the personnel area was calculated as a ratio
(070) of that in the exhaust air for each of the 7
days subsequent
to the introduction
of the
animals. Significant differences
between the
means from the various experimental systems
were determined using Student's t-test.
Results
Airborne dust
Prior to the introduction of animals, no particles
1 Jilll or larger were detected in the air from either
the personnel or exhaust side of any of the
experimental systems. In all rooms the number
of particles increased the first day after animals
were introduced but then decreased gradually
over each successive day. Figure 2 shows the
average number of 0'5, 1'0,2'0,
and 5'0/!m
Airborne bacteria
Figure 3 shows the numbers of airborne bacteria
recovered by the pinhole sampler method from
either side of the rack in each 'system both before
and after the introduction
of animals. In the
absence of animals almost no bacteria were
w
OWAF ROOM + OPEN RACK
TAF ROOf1
10000
--'0
:I:
_Z ~
0>
c:>
c:>
Q.
>-
..-1
0:
W
~
'"::': --..
'"w
--'
•...
u
0:
0
Z
•...
U
.100~===~=_
« '"
=== == == === == == ==
«
--'
'"
w
W
Z
Q.
z
o
o
'"
'"
80
80
60
60
qO
qO
20
20
0:
:<
o
z
0,5
2
0,5
5
0
-1 1
5
w
10000
OWAF + CURTA I NS
OWAF + SLIDING
DOORS
10000
140
--'0
•...
u..
120
«
100
<Xl
'"w
--..
~z
u
'"
w
•...
--'
•...
u
0
« '"
'"w
--'
0:
«
W
Z
0:
Q.
o
'"
:<
o
z
0.5
2
PARTICLE
0.5
SIZE
(11m)
2
PARTICLE
SIZE
(11m)
Fig. 2. Airborne particledistribution in the personnelwork
spaceand exhaust air of each of the 4 experimentalanimal
rooms. The points are shown the averages from the data
collectedfrom day 1 to 7 after the introduction of animals.
0, Work spaceair; e, exhaustair; TAF, turbulent air flow;
OWAF, one-way-airflow.
OWAF + OPEN
RACK
5
1
I
0
-1
7
1
3
5
7
OWAF + SLID I NG DOORS
80
60
f...--l
qO
qO
20
20
Q.
z
«
o
<Xl '"
u..
3
"" • w"","'
:I:
z ~
_0>
c:>
C>
>-..-1
Q.
M
100
ROOM
Q.
'" «
u..
TAF
100
0
0
-1 1
5
3
DAYS AFTER
7
INTRODUCTION
OF ANIMALS
-1 1
3
DAYS AFTER
5
INTRODUCTION
OF ANIMALS
Fig. 3. Numberof airbornebacteriarecoveredby the pinhole
sampler method in the work area and exhaust air in the 4
experimental rooms before and after animals were introduced. 0, Work spaceair; e, exhaust air; TAF, turbulent
airflow; OWAF, one-way-airflow.
Downloaded from lan.sagepub.com at PENNSYLVANIA STATE UNIV on March 3, 2014
One-way
airflow
system in animal rooms
11
detected in any of the systems. After their introduction, however, bacterial counts in the working
area in the TAF room increased with time and
showed the same values as those on the exhaust
side until the fourth day. In the working area of
the open OWAF system, the bacterial counts were
lower than those on the exhaust side except on the
5th and 7th days. In contrast, however, the
number of airborne bacteria recovered from the
working area of the OWAF room with curtains or
sliding doors was only 0-2/100 I of air throughout
the 7 days after the introduction of animals.
Figure 4 shows the counts of settled bacteria
obtained from each experimental room and it can
be seen that once again in the absence of animals,
almost none were detected. After animals were
introduced the numbers gradually increased at
about the same rate and to the same level on both
sides of the rack in the TAF room. The situation
was somewhat better in the open type OWAF
room in that work space air was always less
contaminated than that on the exhaust side. With
curtains or sliding doors, however, there was a
dramatic improvement in the state of the air in
the work space with either none or only one
bacterial colony per dish being detected during
each measurement period.
Ammonia concentration
The results of the ammonia measurements are
summarized in Fig. 5. This shows that in all the
housing systems the ammonia levels increased
from the 4th day onwards. It was relatively high
in the TAF and open OWAF rooms, somewhat
lower in the OWAF room with curtains and
remarkably low in the OWAF room with sliding
OWAF + OPEN RACK
TAF ROO~'
100
100
~ 80
80
200
>-
150
: ~ 150
co
E
IS:
UJ
~ 5U
Q.
100
~ '- 100
«
.
to
~
Q
~
~
>>-
w
V>
50
50
a
a
-1 1
! ~ool
3
-11
3
5
7
I
>>-
~
5
7
80
-: 60
60
50
~ 40
40
~ 20
20
:E
«
o
0
5
5
DAYS AFTER INTRODUCTION
OF ANIMALS
DOORS
>
~
50
OWAF + SLIDING
~ 80
~
100
« .
...J
o
3
w
Q.
~ ~
~ 0
OWAF + CURTAINS
200
w
'"0
z
20
OWAF + SLIDING DOORS
150
r;'- 100
0:
~ 20
1
Q
-
~o
:E
o\jAF + CURTAINS
it; ;;: 150
« >-
>
~ LIO
«
5
5
J: 200
o V>
'" -
60
...J
W
7
DAYS AFTER INTRODUCTION
OF ANIMALS
Fig. 4. Number of settled bacteria recovered by the koch
method in the work area and exhaust air in the 4 experimental
rooms before and after animals were introduced. 0, Work
space air; e, exhaust air; TAF. turbulent airflow; OWAF,
one-way-airflow.
a
a
1
3
5
7
DAYS AFTER INTTODUCTION
OF ANIMALS
1
3
5
7
DAYS AFTER INTRODUCTION
OF ANIMALS
Fig. 5. Ammonia concentration in the air of the work area,
the exhaust air and that inside cages in the 4 experimental
rooms after the introduction of animals.
Work space air;
e, exhaust air; "', cage air; TAF. turbulent airflow; OWAF.
°,
one-way-airflow.
Downloaded from lan.sagepub.com at PENNSYLVANIA STATE UNIV on March 3, 2014
12
Yamauchi,
doors. The low cage levels in the rooms
with curtains or sliding doors are probably
related to the good ventilation within the
cages due to the air control plates. The concentration in the working areas was almost the
same as that on the exhaust side in the TAF room
and about half that on the exhaust side in the
open OWAF room; in both these, however, it
reached at least 20 ppm on days 6 and 7. In
contrast .no ammonia at all was detected in the
working area of the OWAF rooms with curtains
or sliding doors.
Effect of cage cleaning on air cleanliness
in the 0 WAF room with sliding doors
Table
1 shows
the number
of
o· 5 /Lm
Obara,
Fukuyama,
Veda
Airborne bacterial counts and ammonia levels
showed similar, though less massive changes,
peaking during the cage cleaning process but
returning to their preworking levels within 15min
of work ceasing.
Although similar trends occurred on the
exhaust side, the situation was consistently and
very significantly worse; initial particle counts
were around 14oo0/ft3, increased to more than
140000/ft3 on termination of the cleaning procedures and returned to the precleaning level
60 min later. As might be expected, airborne
bacteria and ammonia levels in the exhaust air
were highest before cleaning began and decreased
gradually therafter.
particles,
bacterial counts and ammonia levels detected in
the air on either side of the rack housed in the
OW AF room with sliding doors before, during
and after cage cleaning. Before cleaning commenced the air in the personnel area was below
Class 100. As work began, however, it increased
rapidly to about Class 8000. Peak contamination
(about Class 14000) was reached immediately
after cage cleaning ceased and the air then
became rapidly cleaner, returning to below Class
100 within 30 minutes.
Comparison of air cleanliness in the personnel
area and exhaust side of the animal rooms
Table 2 shows the average contamination levels
of the air in the working areas as a percentage
of that in the exhaust air for each of the 4
experimental set-ups. These ratios were calculated from data obtained during the 7 days
after animals were introduced. The overall
averages show that whereas the contamination
levels in the work space air of the TAF room
reached 90,5070of that of the exhaust air, in the
Table 1. Numbers of airborne dust particles, bacteria and ammonia levels in the personnel working area and exhaust air
in a one-way-airflow room with sliding doors before and after cage changing
0·5 /Lm particles
(No.lftl)
Airborne bacteria
(No.llOO I)
Ammonia
Measuring point
Time
Personnel working area
Before
During
At end
15 min
30 min
45 min
60 min
work
work
of work
later
later
later
later
57'
7986
13 848
3483
57
57
57
Ib
456
25
1
1
0
0
O'Ob
9'6
0'7
0'1
0'0
0,0
0'0
Exhaust air
Before
During
At end
15 min
30 min
45 min
60 min
work
work
of work
later
later
later
later
13 962
53213
142336
74198
37 128
30642
10 478
918
697
307
153
164
44
36
65'2
62'2
12·2
3'7
2·9
1· 5
1·7
(ppm)
'The value shown is the average of the number of particles measured continually 7 times every 2 min through 15 min, except
the single reading at the end of work period.
bThe value shown is the average of 2 measurements.
Downloaded from lan.sagepub.com at PENNSYLVANIA STATE UNIV on March 3, 2014
One-way airflow system in animal rooms
13
Table 2. Average contamination levels of the air in the personnel working area as a percentage of that in the exhaust air
in each of the 4 systems after the introduction of animals
One-way-airflow (OWAF) system
Number of
measurements
(n)
Item
Dust particle size
Airborne bacteria by
pinhole sampler
Settled bacteria by
Koch method
Ammonia
Total
O'31!m
0'51!m
l'Ol!m
2'Ol!m
5'01!m
7
7
7
7
7
Conventional
(TAF) system
Open type
With curtains
(070 )
(Olo)
(Olo)
With sliding
doors (Olo)
29'9±8'3
8·0±2·8
3'2± 1·7
2'4±2'4
0
IO'2±4'1
1·5±0·7
0
0
0
92'9± 7, 58
94'0± 6,3
98'0± 8·1
93'8 ± 8,4
88'0±20'8
7
lOO'7± 18·0
7
7
81·7±11·1
75'2± 13,4
90'5±
56
4,3
52'3±
46'3±
46'3±
48'8±
46'4±
5,9
5·1
6·2
7,8
7,8
49'5± 19'8
13'0±7'2
4'0±3'5
38'4±
30'3±
7,3
8,8
0'2±O'1
l'0±O'8
2·2±1·4
1'6±1'0
46'7±
3' 5b
7'2± 1,9'
2'4±O'8d
n, number of days each item was measured.
8The values were calculated from the 7 daily measurements made under rearing conditions and shown as the mean±SEM
(ratio = value on the working area/the value on the exhaust side x 100).
bSignificant difference by t-test at P<O·OI between conventional and open OWAF systems.
'Significant difference by t-test at P<O·OI between open and curtain type OWAF systems.
dSignificant difference by t-test at P<0·05 between curtain and sliding door OWAF systems.
open OWAF system this was reduced to 46'7OJo;
with curtains added this was further reduced to
7·2% and with sliding doors to only 2·4%.
Discussion
In recent years, accidents caused by zoonoses and
the apparent increase in the incidence of allergies
in scientists and technicians working in close
contact with laboratory animals have become a
problem. In one extreme case a scientist was
infected with haemorrhagic fever with renal
syndrome after being in a rat room for only
10 min (Maejima et al., 1986). According to a
survey on laboratory animal induced allergies
performed by the authors (Study Group on
Laboratory Animal Allergy, 1987),20% of those
with allergies answered that rhinitis, conjunctivitis or asthma occurred after they had only
entered animal rooms.
In almost all conventional animal rooms,
however, a TAF system is used and the open
racks are generally arranged against the walls
with no covering in front. Although in such
rooms a high class of cleanliness (Class 100) can
be maintained before the introduction of animals,
once animals are introduced particle counts rise
markedly to about Class 10 000. Investigations
of airflow patterns in conventional rooms by
means of smoke (Yamauchi et al., 1986) showed
that some of the air from the cages flowed back
to the centre of the room; even some air which
flowed through the cages then rose between the
racks and walls and again returned to the centre
of the room. As would be expected, therefore,
with such TAF systems airborne particles,
bacteria and ammonia are quickly distributed
throughout the room air. This is confIrmed by the
data in Table 2 showing that the air in the work
space soon reached 91% of the contamination
level of that on the exhaust side.
The numbers of airborne particles in the rooms
increased when animals were first introduced
and then decreased with time. This pattern is
related partIy to the behaviour of the animals
which become more active with the environmental
changes and partly to the initially dry bedding
becoming damp. However, the numbers of
airborne bacteria, settled bacteria and ammonia
Downloaded from lan.sagepub.com at PENNSYLVANIA STATE UNIV on March 3, 2014
]4
Yamauchi, Obara, Fukuyama, Veda
levels all increased from about the 4th day
after the introduction of animals. This is
probably related to the accumulation of faeces
and urine excreted by the animals and the
proliferation of bacteria in the cages. These
results suggest that the cages should be replaced
about twice a week.
As far as air cleanliness during cage changing
is concerned, particle counts, bacterial counts
and ammonia levels initially increased markedly
in the work space but returned to the preworking levels within 30 min of the work being
completed. In this situation it has become
necessary for workers to change into special
clothing and to wear a face mask or respirator
and gloves when they work in animal rooms as
has been the case to date. In special cases they
may need to use a safety cabinet when cages are
being changed. According to a study made in a
conventional mouse room there was no correlation between counts of particles and bacterial
counts on the first day after animals were
introduced but there was a significant correlation
on the 3rd and 5th days (Obara et al., 1986).
Such significant correlations between different
parameters will be useful for evaluating the
cleanliness of the working area in animal rooms.
Edwards et al. (1983) found a relationship
between environmental factors and allergens in
animal rooms and reported that in such rooms
with adequate air conditioning systems the
quantity of allergen in the air of the working area
decreased from the rear of the cages. Using their
data it has been calculated that the quantity of
allergen in the work area was 30-850/0 of that
in the exhaust air. These contamination levels
resemble the results obtained here with an
OWAF room with an open rack. It has also been
reported that the quantity of allergen increases
with stocking density, when work is in progress
(Lincoln et 01., 1974; Twiggs et 01., 1982) and
with a reduction in ventilation rates or humidity
levels (Edwards et al., 1983). There are reports
that most of the urinary allergens of guineapigs
are included in dust particles of 5 /Lm or larger
(Swanson et 01., 1984) and most of the urinary
allergens of rats are obtained from dust particles
of at least 7/Lm (Platts-Mills et 01., 1986).
We did not measure the allergens in the air but
while particles of l/Lm or larger were detected
in the TAF and OWAF rooms with an open
rack, almost no particles of such size were
detected in the OW AF rooms with curtains or
sliding doors. This appears to be due to the good
directional control of airflow in the OW AF
rooms. Up to the present the method most
commonly used to maintain air cleanliness in
animal rooms has been to increase the number
of air changes. Indeed, according to the guidelines pertaining to more than 25 countries the
number of air changes recommended for fully
stocked rooms should be between 8 and 20/h
(Clough, 1988). The air change rates employed in
offices and ordinary hospital rooms, however, are
less than 6-8/h (Miyazawa, 1987). In the OWAF
room in this experiment the supply air volume
was set to provide the equivalent of 8 room air
changes/h; yet in relation to the volume of the
rack this was equivalent to 53 air changes/h as
confirmed by measurements of air volumes
leaving from the exhaust slits. In spite of this the
percentage contamination rate of the air in the
work space compared to that of the exhaust air
was maintained at 47% with an open rack, 7%
with curtains and only 2% in the room with
sliding doors. This indicates that the clean air is
utilized very effectively in an OW AF room. The
resulting decrease in overall air change rates also
serves to considerably reduce energy costs
associated with the operation of the air conditioning system.
In conclusion it is suggested that a combination of airflow control and covering the
front of the rack and cages is necessary to
maintain air cleanliness in animal rooms. The
OWAF system used in this experiment can create
a smell-free environment and a very good
standard of cleanliness for the personnel working
in it.
Acknowledgements
This work was supported by Grant-in-aid for
Scientific Research, 60480486 (1985-1987)
from the Ministry of Education, Science and
Downloaded from lan.sagepub.com at PENNSYLVANIA STATE UNIV on March 3, 2014
One-way
airflow
system in animal rooms
Culture of Japan; it was also supported by the
Okumuragumi Co. and the Seiken Co. The
authors wish to acknowledge the technical assistance provided by members of the Institute. We are
15
also grateful for the assistance of Dr G Clough
of Alanann Consultancy Services for his critical
reading of an earlier draft and amendments made
to it.
References
Agrup G, Belin L, Sjostedt L & Skerfving S (1986) Allergy
to laboratory animals in animal technicians and animal
keepers. British Journal of Industrial Medicine 43, 192-198
Botham PA & Teasdale EL (1987) Allergy to laboratory
animals. Biologist 34, 162-163
Clough G (1988) Housing and welfare of laboratory rodents.
In New Developments in Biosciences: Their Implications
for Laboratory Animal Science, pp. 239-244 (eds AC
Beynen & HA Solleveld). Dordrecht: Martinus Nijhoff
Cockcroft A, Edwards J, McCarthy P & Anderson N (1981)
Allergy in laboratory animal workers. Lancet i, 827-830
Committee on Standards of Laboratory Animal Facilities
(1982) Guideline for the construction and equipment of
laboratory animal facilities in 1982, pp. 45-65. Tokyo:
Seishi Shoin (in Japanese)
Davies GE & McArdle LA (1981) Allergy to laboratory
animals: a survey by questionnaire. International Archive
of Allergy and Applied Immunology 64, 302-307
Edwards RG, Beeson MF & Dewdney JM (1983). Laboratory
animal allergy, the measurement of airborne urinary
allergens and the effects of different environmental
conditions. Laboratory Animals 17, 235-239
Federal Standard 209B (1973) Federal Standard Clean Room
and Work Station Requirements, Controlled Environment.
GSA, Printed Materials Supply Division, Building 197,
Naval Weapons Plant, Washington, DC 20407, USA
Gross, NJ (1980). Allergy to laboratory animals. Epidemiologic, clinical, and physiologic aspects, and a trial of
cromolyn in its management. Journal Allergy of Clinical
Immunology 66, 158-165
HMSO (1981) Occupational Asthma 1981. Report by
Industrial Injuries Advisory Council, Command 8121.
London: HMSO
Lincoln TA, Bolton NE & Garrett AS Jr (1974) Occupational
allergy to animal dander and sera. Journal of Occupational
Medicine 16, 465-469
Lutsky II & Neuman I (1975) Laboratory animal dander
allergy. I. An occupational disease. Annals of Allergy 35,
201-205
Maejima K, Ueda K & Yamauchi T (1986) Countermeasure
for hemorrhagic fever with renal syndrome. Laboratory
Animals 3, 3-27 (in Japanese)
Miyazawa H (1987) Planning of air-conditioning in modern
hospitals. Air Conditioning and Refrigeration 27, 53-62
(in Japanese)
Obara T, Fukuyama N, Minakami K, Ueda T & Yamauchi
C (1986) Circadin rhythm of air-borne bacteria and dust
particles in a mouse breeding room. Experimental Animals
35, 357-361 (in Japanese)
Platts-Mills TAE, Heymann PW, Longbottom JL & Wilkins
SR (1986) Airborne allergens associated with asthma,
particle sizes carrying dust mite and rat allergens measured
with a cascade impactor. Journal Allergy of Clinical
Immunology 77, 850-857
Slovak AJM & Hill RN (1981) Laboratory animal allergy,
a clinical survey of an exposed population. British Journal
of Industrial Medicine 38, 38-41
Society for Laboratory Animal Science (1980) Planning
Structures and Construction of Animal Facilities for
Institutes Performing Animal Experiments. Basel: GVSaLAS
Study Group on Laboratory Animal Allergy (1987). Survey
of laboratory animal allergies in Japan. Experimental
Animals 36, 95-101 (in Japanese)
Swanson MC, Agarwal MK, Yunginger JW & Reed CE
(1984) Clinicoimmunologic studies, characterization, airborne quantitation, and size distribution, guinea-pigderived allergens. American Review of Respiratory Disease
129, 844-849
Twiggs JT, Agarwal MK, Bahlberg MJE & Yunginger JW
(1982) Immunochemical measurement of airborne mouse
allergens in a laboratory animal facility. Journal Allergy
of Clinical Immunology 69, 522-526
Yamauchi C, Obara T, Fukuyama N, Ueda T, Enokita D
& Yanagimoto S (1986) One-way air current system
from air supply. side to exhaust side in laboratory
animal rooms. Experimental Animals 35, 537-544 (in
Japanese)
Downloaded from lan.sagepub.com at PENNSYLVANIA STATE UNIV on March 3, 2014