Carbon monoxide euthanasia of dogs: chamber concentrations and comparative effects of automobile engine exhaust and carbon monoxide from a cylinder


Moreland, A.F.

Journal of the American Veterinary Medical Association 165(9): 853-855

1974


Carbon monoxide (CO) produced by combustion in a gasoline engine was compared with CO in commercial cylinders for effectiveness and cost of euthanasia. Differences in effectiveness were considered negligible. Per animal operational costs were less for the gasoline engine method than for the method utilizing commercial cylinders. Initial equipment costs were higher for the gasoline engine than for the commercial cylinder method.

Carbon
Monoxide
Euthanasia
of
Dogs:
Chamber
Concentrations
and
Comparative
Effects
of
Automobile
Engine
Exhaust
and
Carbon
Monoxide
from
a
Cylinder
A.
E
Moreland,
DVM
SUMMARY
Carbon
monoxide
(CO)
produced
by
combustion
in
a
gasoline
engine
was
compared
with
CO
in
commercial
cyl-
inders
for
effectiveness
and
cost
of
euthanasia.
Differences
in
effectiveness
were
considered
negligible.
Per
animal
op-
erational
costs
were
less
for
the
gasoline
engine
method
than
for
the
method
utilizing
commercial
cylinders.
Initial
equipment
costs
were
higher
for
the
gasoline
engine
than
for
the
commercial
cylinder
method.
Destruction
of
animal
life
is
an
offensive
act
to
some
persons.
Veterinarians
spend
much
time
in
training,
learn-
ing
to
prevent
animal
deaths
and
to
ease
animal
distress
and
suffering.
To
a
Doctor
of
Veterinary
Medicine,
trained
in
the
healing
art,
the
idea
of
destroying
animals
is
not
pleasant.
Realizing
that
veterinarians
have
a
responsibility
to
alleviate
distress
and
suffering,
however,
one
concludes
that
these
goals
could
not
be
achieved
without
a
method
of
animal
population
control
that
accomplishes
destruction
of
diseased,
abandoned,
and
unwanted
animals.
Review
of
Literature
Euthanasia
may
be
defined
as
the
act
of
inducing
painless
death
but,
because
the
definition
of
"painless"
death
is
difficult,
the
American
Veterinary
Medical
Association
has
appointed
Pan-
els
on
Euthanasia
to
study
methods
of
euthanasia
in
use
for
un-
wanted
small
animals
and
to
render
reports
suitable
for
publication.
A
variety
of
methods
and
techniques
were
described
by
the
panels.
1,2
One
method
mentioned
was
the
use
of
CO
gas,
but
the
panels
indicated
a
lack
of
sufficient
research
studies
on
the
sub-
ject
and
suggested
additional
carefully
controlled
experiments
di-
rected
toward
standardization
of
procedures;
hence,
the
studies
reported
here
were
undertaken.
When
a
living
animal
is
brought
into
an
environment
contain-
ing
CO,
respiration
brings
the
gas
into
contact
with
the
blood
by
transpulmonary
alveolar
diffusion
and
it
unites
with
hemoglobin,
forming
CO
hemoglobin.
In
this
process,
CO
displaces
0
2
on
the
hemoglobin
molecule,
resulting
in
anoxemia.
Further,
it
is
known
that
the
development
of
poisoning
from
CO
depends
on
the
con-
centration
of
CO
in
the
area
and
the
time
of
exposure.
Reports
on
the
use
of
CO
have
appeared,
4,6,7
but
none
have
Dr.
Moreland
is
Director
of
the
J.
H.
Miller
Health
Center
Animal
Resources
Department,
and
Head,
Division
of
Comparative
Medicine,
College
of
Medicine,
University
of
Florida,
Gainesville,
FL
32601.
Sponsored
by
the
State
of
Florida
and
a
grant
from
the
Greater
Daytona
Dog
Fanciers
Associa-
tion.
The
author
thanks
Dr.
Vernon
P.
Roan,
Mr.
Bill
Buzzell,
and
Manny
Pandreas
of
the
College
of
E
ngineering
for
loan
of
the
gas
analyzer
and
technical
assistance.
dealt
specifically
with
its
use
in
humane
destruction
of
dogs
when
chamber
concentrations
were
accurately
monitored
during
CO
generation
by
combustion
of
gasoline
in
an
engine.
One
unpub-
li
shed
study
compared
methods
of
CO
generation
and
limited
chamber
concentrations.'
Materials
and
Methods
These
studies
compared
the
effectiveness
and
cost
of
delivering
CO
from
commercially
available
prepared
cylinders
and
genera-
tion
of
CO
from
combustion
of
gasoline
in
an
engine.
Further,
the
effect
of
2
different
fl
ow
rates
on
the
efficiency
of
gas
delivered
from
prepared
cylinders
was
studied.
A
lethal
chamber
was
constructed
of
galvanized
sheet
metal
and
measured
24.38
dm
long
by
17.02
dm
wide
by
11.43
dm
high,
having
a
volume
of
4,743
L.
It
was
equipped
with
large
doors
at
both
ends,
a
20-
by
30
-cm
plexiglas
viewing
port,
a
150-w
incan-
descent
bulb
mounted
in
a
protective
grid
inside
the
chamber,
and
a
temperature
sensor
mounted
in
a
midchamber
location.
A
vent
with
a
variable
-sized
opening
was
located
at
one
of
the
upper
corners.
Gas
was
let
into
the
chamber
from
a
port
25
cm
below
the
top
of
the
chamber,
near
the
corner
diagonally
opposite
the
vent.
In
the
studies
utilizing
cylinder
gas
at
a
fl
ow
rate
of
15
L/minute,
the
gas
was
introduced
through
fl
exible
tubing
placed
through
the
port
and
piped
to
the
4
corners
of
the
chamber,
within
8
cm
of
the
chamber
fl
oor.
A
1962
Dodge
V8,
318
-cubic
-inch
-displacement
gasoline
engine
was
removed
from
a
wrecked
vehicle
and
installed
in
a
specially
prepared
permanent
angle
-iron
frame
alongside
the
euthanasia
chamber.
The
engine
cost
$75
and
the
estimated
cost
of
materials
and
labor
to
construct
the
angle
-iron
frame
and
to
complete
the
installation
made
a
total
cost
of
approximately
$500.
Exhaust
gas
from
the
engine
was
first
piped
into
a
55
-gal
steel
drum
through
a
5.08
-cm
pipe
(ID)
fitted
into
the
drum
7.5
dm
above
the
bottom.
As
the
inlet
pipe
passed
into
the
drum,
it
was
turned
downward
by
an
angle
of
90
degrees,
terminating
approxi-
mately
5
cm
above
the
bottom
of
the
drum.
Water
was
put
into
the
drum
to
a
depth
of
about
6
dm,
so
that
the
gas
would
be
cooled
and
cleansed
prior
to
collecting
above
the
water
and
pass-
ing
through
an
outlet
pipe
fitted
near
the
top
of
the
drum.
The
gas
was
then
piped
to
and
filtered
through
a
7.62-
by
20.32-
by
25.4
-cm
chamber
to
filter
carbon
particles
from
the
gas.
Gauze
was
placed
over
the
inlet
and
outlet
pipes
and
the
filter
chamber
was
filled
with
silica
gel
to
prevent
humidity
from
the
water
filter
from
condensing
on
the
wall
of
the
sample
line
and
on
filters
in
the
gas
analyzer.
A
sensor
probe
from
an
indoor
-outdoor
ther-
mometer
was
installed
in
the
filter
chamber.
This
sensor
and
the
one
monitoring
temperature
in
the
interior
of
the
lethal
chamber
led
to
a
thermometer
console
mounted
on
the
outside
of
the
le-
thal
chamber,
in
plain
view
of
the
engine
operator.
Chamber
tem-
perature
did
not
exceed
90
F.
Concentration
of
gases
within
the
lethal
chamber
was
mea-
sured
by
a
gas
analyzer,a
which
utilizes
the
principle
of
infrared
a
MSA
Lira
300,
Mine
Safety
Appliances
Company,
Inc,
Instrument
Division,
Pittsburgh,
PA.
November
1,
1974
853
absorption
spectroscopy
and
is
designed
to
measure
CO
concen-
trations
from
0
to
10%,
nitric
oxide
from
0
to
5,000
parts
per
million
(ppm),
and
hydrocarbons
from
0
to
1,500
ppm.
Prior
to
1970,
measurements
of
gaseous
pollutants
were
largely
made
by
modifications
of
wet
chemical
procedures,
which
gave
marginal
sensitivity
and
specificity.
Spectroscopic
techniques
offer
means
for
the
direct
and
continuous
detection
of
the
pollutant
in
the
gas
phase.
5
It
is
for
this
reason
and
for
improved
accuracy
that
this
instrument
was
chosen
to
determine
concentrations
of
engine
ex-
haust
pollutants.
Cylinder
gas
and
regulator
used
in
these
experiments
are
com-
mercially
available.
b
The
CO
(99.5%
pure)
was
obtained
in
a
size
1A
cylinder.
Flow
of
gas
was
regulated
by
a
2
-stage
regulator
that
allowed
delivery
at
a
pressure
4
to
50
psi.
In
3
trials
with
this
gas,
a
locally
fabricated
fl
ow
meter
measuring
a
fl
ow
rate
of
15
L/
minute
was
utilized.
In
2
trials,
a
rotametere
that
indicated
a
fl
ow
rate
of
100
L/minute
for
2
minutes
was
used
and
in
2
trials,
the
same
rotameter
was
used
for
3
minutes.
Dogs
of
both
sexes
and
mixed
breeding
were
obtained
from
municipal
animal
pounds.
Ages
varied
from
young
adult
to
ap-
proximately
10
years
of
age
and
weights
varied
from
5.4
to
29.0
kg.
Groups
of
dogs
were
randomly
selected
to
simulate
actual
field
conditions.
Trials
1,
2,
3,
4,
5,
and
6,
using
2,
16,
5, 5,
16,
and
7
dogs,
respectively,
were
conducted
with
the
gasoline
engine
idling
on
maximum
choke.
Trials
7,
8,
and
9,
with
9, 9,
and
16
dogs,
re-
spectively,
were
done
with
CO
from
the
cylinder,
at
a
fl
ow
rate
of
15
L/minute.
Trials
10
and
11,
with
5
and
7
dogs,
respectively,
were
done
with
cylinder
CO
fl
owing
through
the
rotameter
at
a
fl
ow
rate
of
100
L/minute
for
2
minutes.
In
trials
12
and
13,
with
6
dogs
each,
the
rotameter
was
used
to
provide
a
fl
ow
rate
of
100
L/minute
for
3
minutes.
Gases
other
than
CO
(nitric
oxide
and
hydrocarbons)
that
are
produced
by
the
process
of
combustion
in
a
gasoline
engine
are
thought
by
some
to
occur
in
sufficiently
high
concentration
to
cause
irritating
effects
to
eyes
and
respiratory
mucous
mem-
branes
when
this
method
of
CO
production
is
used.
Thus,
these
factors
were
measured
in
trials
1
through
6.
Results
and
Discussion
All
of
the
methods
of
CO
delivery
appeared
effective.
Less
than
1.3
minutes'
variation
occurred
in
the
average
total
time
elapsed
between
initial
exposure
and
death
of
the
last
dog
when
methods
of
delivery
were
compared
(Ta-
ble
1).
Other
events
(Table
1)
occurred
with
little
more
than
1
to
1
1
/
2
minutes'
variation
between
the
methods.
Only
one
trend
appears
obvious,
namely,
the
interval
be-
tween
initial
exposure
and
first
collapse
seems
correlated
with
speed
of
administration
of
the
gas.
Differences
be-
tween
the
gasoline
engine
and
cylinder
gas
at
a
fl
ow
rate
of
15
L
/minute
were
negligible.
Indeed,
if
a
fl
ow
rate
for
CO
in
the
gasoline
engine
exhaust
were
calculated
on
the
basis
of
CO
concentrations
compared
with
time,
a
CO
fl
ow
rate
of
approximately
18
L
/minute
is
obtained.
These
data
are
in
accord
with
the
findings
of
others'
who
reported
a
lesser
initiation
-collapse
interval
when
dogs
were
placed
through
a
top
door
into
a
chamber
already
charged
to
a
concentration
of
6%
CO.
Because
of
limita-
tions
inherent
in
equipment,
however,
the
interval
be-
tween
initiation
of
fl
ow
and
achievement
of
6%
CO
concentration
could
not
be
reduced
in
the
present
studies
to
less
than
3
minutes,
the
limiting
factor
apparently
being
the
pressure
range
of
the
cylinder
pressure
-regulator
valve.
All
of
the
aforementioned
methods
show
a
marked
im-
provement
in
the
average
time
of
death,
compared
with
a
b
The
Matheson
Company,
Gas
Products
Division
of
Will
Ross,
Inc,
East
Rutherford,
NJ.
e
Brooks
Instrument
Company,
Inc,
Hatfield,
PA.
sulfuric
acid
-sodium
formate
reaction
method
of
CO
gener-
ation.'
Further,
it
would
appear
that
inasmuch
as
signs
of
significant
anxiety
or
distress
prior
to
collapse
were
not
observed
in
the
present
experiments,
all
methods
tested
would
be
acceptable
for
mass
euthanasia.
It
is
probably
possible
to
shorten
the
times
observed
in
these
experi-
ments
by
either
reducing
the
volume
of
the
lethal
chamber
or,
in
the
case
of
use
of
cylinder
gas,
by
using
a
cylinder
pressure
-regulator
valve
of
higher
pressure
range.
Howev-
er,
it
is
problematic
whether
the
estimated
further
reduc-
tion
of
1
to
3
minutes
in
average
death
time
in
this
large
chamber
would
provide
a
significant
advantage.
The
best
way
to
reduce
the
time
required
in
a
large
chamber
is
to
charge
the
lethal
chamber
prior
to
placement
of
animals
within
it.
As
already
mentioned,
this
technique
requires
top
loading
and
is
aesthetically
unacceptable
to
many
per-
sons
because
of
the
piling
up
of
dead
animals.
Nitric
oxide
concentration
was
less
than
100
ppm
throughout
all
trials
(Table
2)
and
since
the
analyzer
scale
measures
such
a
wide
range
(0
to
5,000
ppm),
accuracy
in
reading
the
scale
at
these
low
levels
was
difficult.
In
cases
of
doubt,
however,
scale
readings
were
always
made
on
the
high
side.
Concentration
of
hydrocarbons
did
not
exceed
375
ppm
at
any
time
in
any
trial,
the
highest
average
concentration
being
245
ppm
(Table
2).
If
the
concentration
was
recorded
at
the
average
time
of
collapse,
i.e.,
4.3
minutes,
which
ap-
proximates
the
time
at
which
all
dogs
were
unconscious,
the
average
concentration
was
approximately
130
ppm.
Lacrimation,
sneezing,
or
coughing,
usual
signs
of
ocular
and
respiratory
irritation,
were
not
noticed.
Environmen-
tal
engineers
at
the
University
of
Florida
speculate
that
these
levels
of
nitric
oxide
and
hydrocarbons
are
too
low
to
cause
significant
biologic
effects
during
these
brief
expo-
sure
intervals;
however,
conclusive
data
to
support
this
opinion
are
unavailable.
The
lethal
chamber
used
for
these
experiments
was
easy
to
load
and
unload,
and
to
sanitize
following
use.
The
prin-
cipal
disadvantage
of
this
chamber
(compared
with
a
top
-
loading
chamber)
lies
in
the
necessity
for
complete
replace-
ment
of
the
gas
each
time
the
chamber
is
loaded,
thus
making
the
process
slower
and
increasing
the
cost
of
opera-
tion
per
animal.
Despite
this
disadvantage,
however,
this
type
chamber
is
better
than
the
top
-loading
chamber
be-
cause
of
its
greater
aesthetic
acceptability.
The
fl
oor
area
of
the
chamber
gave
sufficient
space
for
as
many
as
25
large
dogs
to
stand
erect
comfortably,
and
its
height
facili
-
tated
entry
by
attendants
for
the
purpose
of
removal
of
the
animals
and
subsequent
sanitization
of
the
compartment.
Of
course,
with
both
types
of
chambers,
it
is
imperative
that
the
chamber
be
located
in
a
well
-ventilated
space,
preferably
outdoors.
Reaction
of
dogs
to
the
gas
was
essentially
the
same
fo
r
all
methods
tested.
First,
there
was
a
short
period
during
which
dogs
continued
to
move
about,
wag
their
tails,
or
even
quarrel
with
one
another.
Later,
muscle
quivering,
weakness,
and
ataxia
occurred,
followed
suddenly
by
com
-
plete
collapse
and
apparent
unconsciousness.
Collapse
wa
s
followed
by
a
15-
to
20
-second
period
of
involuntary
exci
te-
ment
—pedaling
motions
of
the
limbs
accompanied
bY,
crying
or
whimpering.
The
stage
of
involuntary
exciterneo,
was
followed
by
slow,
irregular,
labored
respiration,
termi-
nating
:
in
a
series
of
gasps,
with
increasing
intervals
be
,
tween
each.
Usually
at
this
stage
a
violent
final
respira
tory
854
JAVMA,
Vol
165,
No.
9
TABLE
1
-Comparison
Between
Methods
of
Supplying
Carbon
Monoxide
(CO)
and
Times
of
First
and
Last
Collapse,
First
Death,
Last
Death,
and
Total
Exposure
Average
time
from
exposure
to
fi
rst
collapse
(min);
CO
conc.
at
collapse§
Average
time
from
fi
rst
collapse
to
last
collapse
(min);
CO
cone.
at
last
collapse§
Average
time
from
fi
rst
collapse
to
fi
rst
death
(min)
;
CO
conc.
at
fi
rst
death§
Average
time
from
fi
rst
to
last
death
(min);
CO
conc.
at
last
death§
Average
time
from
fi
rst
collapse
to
last
death
(min)
Average
time
from
exposure
to
last
death
(min)
Gasoline
engine
3.20
1.5
2.70
4.00
6.70
9.92
idling
and
at
full
choke*
(0.91)
(1.5)
(1.77)
(2.29)
Cylinder
gas
-
2.93
1.9
2.31
5.67
7.98
10.91
15
L/minute**
(0.89)
(1.3)
(1.33)
(1.86)
Cylinder
gas
-
100
L/minute
for
2.20
1.3
3.86
4.69
8.55
10.75
2
minutest
(3.50)
(4.2)
(4.2)
(4.2)
Cylinder
gas
-
2.00
3.5
3.50
4.12
7.62
9.62
100
L/minute
for
(4.10)
(6.2)
(6.2)
(6.2)
3
minutest
Six
trials,
using
2,
16,
5,
5,
16,
and
7
dogs.
**
Three
trials,
using
9,
9,
and
16
dogs.
t
Two
trials,
using
5
and
7
clogs.
$
Two
trials,
using
6
dogs
each.
§
Values
in
parentheses
=
percentage
of
CO
concentration.
TABLE
2
-Average
Concentrations
of
Nitric
Oxide
and
Hydrocarbons
TABLE
3
-Cost
of
CO
Cylinder
Gas
at
June,
1974,
Price
per
Size
During
Production
of
CO
by
Internal
Combustion
Engine
1A
Cylinder
Used
in
a
4,743-L
Lethal
Chamber
Time
(min)
Nitric
oxide
(ppm)
Hydrocarbons
(ppm)
0
0
0
1
45
17.5
2
60
52.4
3
50
92.0
4
70
128.6
5
70
150.8
6
65
198.6
7
80
234.4
8
60
245.0
9
55
221.6
10
55
227.5
11
55
234.1
12
55
238.3
13
48
240.0
effort
and
tonic
muscular
spasm
occurred,
after
which
body
sphincters
relaxed
and
the
dogs
urinated,
defecated,
and
occasionally
regurgitated.
Death
was
judged
to
have
occurred
when
respiration
ceased,
although
in
all
instances
dogs
were
left
under
observation
within
the
chamber
for
at
least
5
additional
minutes.
On
opening
both
chamber
doors
and
after
waiting
2
or
3
minutes
to
allow
dispersal
of
the
gas,
dogs
were
removed
and
cessation
of
heartbeat
was
ver-
ified
by
auscultation.
Although
the
initial
cost
of
the
gasoline
engine
installa-
tion
is
higher,
operational
costs
of
that
system
are
lower
than
a
system
using
cylinder
gas.
A
service
vehicle
engine
would
not
be
effective
because
most
such
vehicles
are
equipped
with
automatic
chokes,
which
effectively
pre-
clude
maximal
CO
generation.
Late
-model
vehicles
are
made
even
less
effective
because
of
US
Government
re-
quirements
for
various
anti
-pollution
devices.
Costs
of
a
chamber
equipped
with
internal
lighting,
ther-
mometers,
and
water
and
gauze
filters
is
essentially
equal
for
engine
-generated
or
cylinder
gas
although
the
filters
can
be
eliminated
for
the
latter.
Purchase
and
installation
cost
of
the
engine
may
vary,
but
in
this
case
was
about
$500.
Costs
for
the
cylinder
gas
system
include
a
pressure
-
regulator
valve
at
about
$58;
rotameter,
about
$70;
cylin-
der
cart,
about
$50;
associated
tubing
and
connectors,
about
$5;
and
labor
for
assembly,
an
estimated
$20,
for
a
total
of
about
$203.
A
cylinder
deposit
of
about
$54
is
a
refundable
expense
of
this
system.
CO,
chemically
pure*
Per
chamber
charge
Per
animal
cost**
2%
$1.027
$0.04108
3%
$1.536
$0.06144
4%
$2.012
$0.08048
5%
$2.509
$0.1003
6%
$3.018
$0.1207
$52.50/cylinder.
**
25
dogs/chamber
charge.
At
a
cost
of
$0.50
to
$0.55/
gal
of
gasoline,
6
to
8
minutes
of
engine
combustion
time
will
usually
amount
to
less
than
$0.10
worth
of
gasoline.
Allowing
generously
for
labor
costs
in
maintenance
of
the
engine,
per
animal
cost
should
be
less
than
$0.01.
Carbon
monoxide
in
a
cylinder
(99.5%
puri-
ty)
currently
costs
$52.50
/cylinder
(4,955
L)
-a
cost
of
$0.01059/L.
To
charge
this
4,743-L
chamber
to
2,
3,
4,
5,
or
6%
with
CO
requires
97,
145,
190,
237,
and
285
L,
respectively.
Though
chemically
pure
CO
was
used
in
these
studies,
the
technical
grade
gas
(which
is
99.0%
pure)
should
be
cheaper
and
virtually
identical
in
effectiveness.
Clearly,
the
use
of
the
gasoline
engine
is
most
economic;
however,
use
of
cyl-
inder
gas
may
be
easier
in
many
situations
and
cost
is
rea-
sonable
(Table
3)
compared
with
that
for
euthanasia
methods
that
require
greater
labor
for
handling
of
animals.
References
1.
AVMA
Council
on
Research:
Council
Report
-Report
of
the
AVMA
Panel
on
Euthanasia.
JAVMA,
142,
(Jan
15,
1963):
162-170.
2.
AVMA
Council
on
Research:
Council
Report
-Report
of
the
AVMA
Panel
on
Euthanasia.
JAVMA,
160,
(March
1,
1972):
761-772.
3.
Blood,
D.
C.,
Johnston,
D.
E.,
and
Blackwood,
J.
D.,
School
of
Veter-
inary
Science,
University
of
Melbourne,
and
the
Division
of
Chemical
Engineering
Commonwealth
Scientific
and
Industrial
Research
Organiza-
tion,
Melbourne,
Australia:
Unpublished
data,
ca
1969-70.
4.
Carding,
A.
H.:
Mass
Euthanasia
of
Dogs
with
Carbon
Monoxide
and/or
Carbon
Dioxide:
Preliminary
Trials.
J
Small
Anim
Prac,
9,
(1968):
245-259.
5.
Hodgeson,
J.
A.,
McClenny,
W.
A.,
and
Hanst,
P.
L.: Air
Pollution
Monitoring
by
Advanced
Spectroscopic
Techniques.
Science,
182,
(1973):
248-258.
6.
Ramsey,
T.
L.,
and
Eilmann,
H.
J.:
Carbon
Monoxide
Acute
and
Chronic
Poisoning
and
Experimental
Studies.
J
Lab
&
Clin
Med,
17,
(1932):
415-427.
7.
Vinter,
F.
J.:
The
Humane
Killing
of
Mink.
Brit
Fur
Farmers'
Ga-
zette,
Aug,
1957.
November
I,
1974
855