Efficacy of two antibiotic treatments in curing clinical and subclinical mastitis in lactating dairy cows


McDougall, S.

New Zealand Veterinary Journal 46(6): 226-232

2005


595 cows with clinical or subclinical mastitis diagnosed between 8 July and 21 August 1997 in the Morrinsville region, New Zealand were used to evaluated the efficacy of 2 antibiotic treatments. 294 cows were treated (intramammary injection) with procaine penicillin at 0.5 g and dihydrostreptomycin at 0.5 g on 3 occasions at 12-h intervals (P-DHS) and 301 cows were treated with 2 s.c. injections of 10 and 5 gof penthamate hydriodide in aqueous solution (PEN) 24 h apart. Milk samples were collected before and 14 and 21 days after initiation of treatment for bacteriological culture, somatic cell count determination and conductivity testing. Streptococcus uberis was commonest isolate followed by coagulase-negative staphylococci, Staphylococcus aureus, Streptococcus dysgalactiae and coliforms. No significant differences in the bacteriological cure rate of major Gram-positive pathogens, clinical cure rate, somatic cell count or conductivity were found between treatments. However, the bacteriological cure rate of coagulase-negative Staphylococcus infections and the overall bacteriological cure rate was lower for quarters treated with PEN than with P-DHS.

226
New
Zealand
Veterinary
Journal
1998
Efficacy
of
two
antibiotic
treatments
in
curing
clinical
and
subclinical
mastitis
in
lactating
dairy
cows
S.
McDougall*
Abstract
Aim.
To
compare
the
bacteriological
and
clinical
cure
rates
for
clinical
and
subclinical
mastitis
in
New
Zealand
dairy
cows
following
treatment
with
either
an
intramammary
penicillin-dihydrosptreptomycin
preparation
or
a
subcutaneous
injection
of
penethamate
hydriodide.
Methods.
Milk
samples
were
collected
from
clinical
and
subclinical
cases
of
mastitis
before
and
14
and
21
days
after
initiation
of
treatment
for
bacteriological
culture,
somatic
cell
count
determination
and
conductivity
testing.
Results.
No
significant
differences
in
the
bacteriological
cure
rate
of
major
Gram-positive
pathogens,
clinical
cure
rate,
somatic
cell
count
or
conductivity
were
found
between
treatments.
However,
the
bacteriological
cure
rate
of
coagulase-negative
staphylococcus
infections
and
the
overall
bacteriological
cure
rate
was
lower
for
quarters
treated
with
penethamate
than
with
penicillin-dihydrosptreptomycin.
Conclusions.
The
bacteriological
cure
rate
of
mastitis
caused
by
major
Gram-positive
pathogens,
the
clinical
cure
rate,
somatic
cell
count
and
conductivity
did
not
differ
between
the
two
antibiotic
treatments.
Key
words.
Clinical
mastitis,
antibiotics,
bacteriological
cure
rate,
somatic
cell
count,
conductivity.
(New
Zealand
Veterinary
Journal
46,
226-232,
1998.)
Introduction
Mastitis
and
associated
elevations
in
somatic
cell
count
(SCC)
result
in
significant
costs
to
New
Zealand
dairy
herd
owners.
It
has
been
estimated
that
a
bulk
milk
SCC
of
400
000
cells/ml
results
in
a
loss
of
$14
000
per
annum
due
to
the
costs
of
treatment,
milk
withholding
during
treatment,
lost
production,
cow
deaths
and
costs
associated
with
cull-
ing
cows
not
responding
to
treatment
(Woolford,
1997).
Clinical
mastitis
is
diagnosed
where
there
is
grossly
vis-
ible
change
in
milk
composition
(clots,
wateriness)
and/or
changes
to
the
mammary
gland
(pain
and/or
swelling).
Subclinical
infection
occurs
where
bacteria
can
be
cultured
from
milk
but
no
clinical
signs
are
detectable.
Treatment
of
clinical
mastitis
during
lactation
is
an
es-
sential
part
of
mastitis
control
programmes
(Morris
et
al.,
1978;
Woolford
et
al.,
1995).
Treatment
aims
to
eradicate
bacteria
from
the
udder,
reduce
the
production
losses
as-
sociated
with
infection,
minimise
damage
to
milk
secre-
tory
tissue
and
alleviate
pain
associated
with
infection
(Cra-
ven,
1987).
The
selection
of
antibiotic(s)
for
treatment
will
depend
on
the
species
of
bacteria
present,
the
sensitiv-
ity
pattern
of
that
bacterial
species,
the
penetration
of
the
antibiotic
to
the
actual
site
of
infection,
minimisation
of
the
chance
of
developing
resistance
and
the
toxicity
to
the
animal
(Whittem
and
Gaon,
1998).
The
efficacy
of
treat-
ment
of
clinical
mastitis
is
assessed
on
the
basis
of
removal
of
clinical
signs
(such
as
swelling,
heat
and
pain
of
the
udder
or
clots
or
blood
in
the
milk)
and
on
removal
of
bacteria
(Anon,
1996).
The
efficacy
of
antibiotic
treatment
of
mastitis
varies
between
bacterial
species,
farms,
cows,
and
the
dose,
duration
and
route
of
administration
of
the
antibiotic
(Craven,
1987).
Bacterial
species
associated
with
mastitis
in
New
Zea-
land
have
been
reported
as
including
Streptococcus
uberis
*
Animal
Health
Centre,
P.O.
Box
21,
Morrinsville,
New
Zealand.
(SU),
Staphylococcus
aureus
(SA),
Streptococcus
4ysgalactiae
(SD),
Streptococcus
agalactiae,
coagulase-negative
staphy-
lococcus
(CNS),
coliforms
and
Corynebacterium
bovis
(Brookbanks,
1966;
Pankey
et
al.,
1982;
Williamson
et
al.,
1995).
Many
new
infections
occur
around
calving,
with
10%
of
cows
acquiring
new
infections
during
the
early
part
of
lactation
(Pankey
et
al.,
1982;
Williamson
et
al.,
1995).
Sixty-seven
percent
of
clinical
cases
in
heifers
within
the
first
5
days
of
lactation
were
due
to
SU
in
one
New
Zealand
study
(Pankey
et
al.,
1982).
Laboratory
submis-
sions
indicate
that
Streptococcus
spp.
from
clinical
mastitis
cases
are
generally
sensitive
to
penicillin
(99.5%
of
iso-
lates;
Roger
Ellison,
pers.
comm.).
An
intramammary
preparation
containing
1
g
of
penicil-
lin
and
0.5
g
of
dihydrostreptomycin
(Streptopen
High
Potency
Milking
Cow,
Schering-Plough
Animal
Health,
Upper
Hutt)
used
three
times
at
12-hour
intervals
is
com-
monly
used
for
the
treatment
of
clinical
mastit:is,
as
it
has
the
highest
concentration
of
penicillin
of
any
of
the
intramammary
preparations
commercially
available
in
New
Zealand.
Penethamate
hydriodide
in
aqueous
solution
(Leo
cillin
injection,
Bo
ehringer
Ingelheim
NZ
Ltd,
Manukau
City)
is
used
as
a
parenteral
treatment
for
masti-
tis.
It
is
a
weak
base
and
is
lipophilic,
which
results
in
high
concentrations
of
penicillin
in
the
mammary
gland
following
intramuscular
injection
(Ziv,
1980).
Penethamate
hydriodide
has
not
been
previously
evaluated
using
a
treat-
ment
regime
of
10
g
(20
mg/kg
liveweight
for
a
500
kg
cow)
followed
24
hours
later
by
5
g
or
following
subcuta-
neous
injection.
Subcutaneous
injection
has
the
advantage
of
minimising
tissue
damage
and
hence
reducing
the
loss
of
revenue
due
to
carcass
trimming.
This
trial
evaluated
the
efficacy
of
treatment
of
clinical
mastitis
using
two
commercially
available
antibiotic
prepa-
rations
under
New
Zealand
field
conditions.
The
specific
hypothesis
being
tested
in
this
study
was
that
there
was
no
difference
between
a
penicillin-dihydrostrepto-
mycin
intramammary
preparation
and
a
parenteral
treatment
with
penethamate
hydriodide
in
the
clinical,
cytological
or
1998
New
Zealand
Veterinary
Journal
227
bacteriological
cure
of
naturally
occurring,
penicillin-sensi-
tive
cases
of
clinical
mastitis
in
New
Zealand
dairy
cows.
The
effect
of
treating
subclinical
cases
of
mastitis
detected
at
the
time
of
diagnosis
of
clinical
mastitis
was
also
examined.
Materials
and
Methods
Herd
owners
(n
=
38)
were
asked
to
present
for
examina-
tion,
within
48
hours,
any
cow
that
was
diagnosed
as
hav-
ing
clinical
mastitis
between
8
July
and
21
August
1997.
Herd
owners
were
selected
on
the
basis
of
being
within
20
km
of
Morrinsville,
New
Zealand
and
their
willingness
to
be
involved
in
the
trial
and
to
follow
the
trial
protocol.
At
presentation
(day
0;
DO)
the
following
procedures
were
performed
by
a
trained
technician:
The
owner
was
asked
to
identify
the
quarter(s)
diag-
nosed
with
clinical
mastitis.
Each
quarter
was
palpated
for
heat,
swelling
and
oedema.
The
conductivity
of
each
quarter
was
assessed
by
plac-
ing
the
first
three
strippings
of
milk
into
the
bowl
of
a
conductivity
meter
(Milk-Checker,
Technipharm,
Rotorua).
The
milk
was
assessed
for
the
presence
of
clots
and/or
blood.
Milk
samples
(20
ml)
were
collected
aseptically
for
bacterial
culture
and
sensitivity
and
for
analysis
of
SCC.
Samples
were
collected
from
all
quarters
identified
as
hav-
ing
clinical
mastitis
by
the
herd
owner
and
any
additional
quarters
where
the
conductivity
was
greater
than
15%
above
the
lowest
quarter
within
the
cow.
These
later
quarters
were
defined
as
having
subclinical
mastitis.
Quarters
from
which
bacteria
were
isolated
at
DO
were
re-examined
and
re-sampled
(for
bacteriology,
SCC
and
conductivity),
be-
tween
13-15
(D14)
and
20-22
(D21)
days
later,
respec-
tively.
Herd
owners
were
instructed
to
re-treat
any
quarter
that
they
perceived
had
not
responded
to
treatment,
12
or
more
hours
after
the
completion
of
the
treatment
course
and
to
record
this
data.
Treatment
was
with
one
of
two
commercially
available,
registered
antibiotic
preparations.
The
first
contained
1
g
of
procaine
penicillin
and
0.5
g
of
dihydrostreptomycin
(P/DHS)
which
was
infused
into
the
mammary
gland
on
three
occasions
at
about
12-hour
intervals
after
the
quar-
ter
was
milked
out.
The
teat
end
was
scrubbed
with
a
cotton
wool
ball
soaked
in
70%
methanol
before
the
an-
tibiotic
was
infused.
The
second
involved
two
subcutane-
ous
injections
24
hours
apart
of
10
g
and
5
g
of
penethamate
hydriodide
in
aqueous
solution
(PEN).
Sub-
cutaneous
injections
were
administered
into
the
anterior
half
of
the
neck
using
a
20
ml
syringe
with
a
16
G
x
25
mm
needle,
using
the alternate
side
of
the
neck
for
the
second
treatment.
The
initial
treatments
were
administered
by
the
trial
tech-
nicians
and
the
second
and/or
third
treatment
were
adminis-
tered
by
the
herd
owner.
The
herd
owners
were
asked
to
discard
the
milk
from
all
four
quarters
for
144
and
48
hours
after
the
final
treatment
of
P/DHS
and
PEN,
respectively.
Cows
were
assigned
to
a
treatment
from
a
pre-prepared
list
of
treatments,
with
a
different
list
prepared
for
each
herd.
The
treatment
list
was
prepared
by
randomly
assign-
ing
one
of
each
treatment
type
to
each
successive
pair
of
cows.
The
treatment
type
was
known
to
the
herd
owner
and
technician
as
the
route
of
delivery
identified
the
treat-
ment.
Where
more
than
one
quarter
within
a
cow
was
dia-
gnosed
as
having
clinical
mastitis
the
same
treatment
was
used
for
all
quarters.
Laboratory
procedures
Milk
samples
were
frozen
(-20
°C)
for
up
to
7
days
before
thawing
and
culture.
Freezing
does
not
reduce
the
probability
that
bacteria
will
be
isolated
(Murdough
et
al.,
1996).
A
10µl
quantity
of
milk
was
removed
from
the
sample
vial
with
a
sterile
loop
and
applied
to
a
quarter
plate
of
5%
sheep
blood
agar
containing
0.1%
aesculin.
Plates
were
assessed
after
24
and
48
hours
incubation
at
37
°C.
Isolates
were
categorised
on
the
basis
of
colony
morphology,
Gram's
stain,
haemolysis
pat-
tern,
aesculin
reaction,
tube
coagulase
and
CAMP
tests.
Coliforms
were
further
cultured
on
MacConkey's
agar
and
had
an
oxidase
test
performed.
Major
Gram-positive
patho-
gens
were
defined
as
SU,
SA
and
SD
and
the
minor
pathogens
included
Actinobacillis
pyogenes,
Bacillis
spp.,
CNS,
Corynebac-
terium
spp.
and
Pseudomonas
aeruginosa.
Sensitivity
to
a
range
of
antibiotics
of
the
majority
of
SA
(n
=
30)
and
a
selection
of
the
SU
(n
=
10)
and
E.
coli
(n
=
12)
isolates
were
tested
using
the
agar
disk
diffusion
(Kirby-Bauer)
method.
The
antibiotics
tested
included
oxacillin,
cephalothin,
tetracycline,
amoxycillin/
clavulanic
acid,
penicillin,
streptomycin,
erythromycin,
septrim
and
triple
sulfa.
Somatic
cell
counts
were
performed
using
a
Fossomatic
360,
420
or
5000
at
the
National
Milk
Analysis
Laboratory
(Livestock
Improvement
Corporation,
Hamilton)
within
1
4
days
of
collection
after
storage
at
4
°C.
The
SCCs
were
divided
by
1000
and
log
transformed
for
analysis.
The
interquarter
ratio
(IQR)
of
conductivity
was
calcu-
lated
as
the
difference
in
conductivity
between
each
quar-
ter
and
the
lowest
quarter
within
the
cow:
_
Quartet;
Quarter,.
IQR(%)
-
x
100(%)
Quarter,.
where:
Quarter
=
the
conductivity
(in
mS/cm)
for
the
quarter
in
question;
and
Quarter
,
=
the
conductivity
(in
mS/cm)
of
the
quarter
with
the
lowest
conductivity
within
a
cow.
The
IQRs
were
log
transformed
(after
addition
of
1
due
to
0
values)
to
normalise
the
data.
Statistical
analysis
Herd
owners
were
asked
not
present
any
cow
that
had
been
treated
previously
in
the
lactation
with
antibiotic
or
anti-inflammatory
preparations
or
any
cow
that
had
inter-
current
disease,
teat
damage
or
that
was
showing
signs
of
systemic
effects
of
mastitis.
Four
cows
were
not
enrolled
following
presentation,
as
they
displayed
systemic
effects
of
illness,
including
elevated
rectal
temperature
(>
40
°C)
and/or
signs
of
depression,
lethargy,
diarrhoea
or
dehy-
dration.
Clinical
quarters
were
only
included
in
analysis
if
there
was
evidence
of
both
cytological
change
(i.e.
SCC
on
DO
was
>
500
000
cells
per
ml)
and
tissue
damage
(i.e.
there
was
clots
or
blood
in
the
milk
or
udder
oedema
was
present;
Anon,
1996).
A
total
of
67
(8.4%)
of
quarters
diagnosed
by
herd
owners
as
having
clinical
mastitis
were
removed
from
analysis
as
the
SCC
was
<
500
000
cells/
ml
and/or
the
technicians
did
not
detected
gross
changes
to
the
milk
or
udder.
There
was
no
difference
in
the
num-
bers
removed
from
the
PEN
treated
group
(n
=
39)
and
the
P/DHS
group
(n
=
28;
x
2
=
1.5,
p
>
0.2).
A
further
184
(23.0%)
quarters
were
removed
from
analysis
due
to
228
New
Zealand
Veterinary
Journal
1998
insufficient
clinical
data
or
because
estimates
of
the
SCC
could
not
be
made
due
to
the
large
number
of
clots
in
the
milk
which
interfered
with
the
counting
process.
The
quarter
was
the
unit
of
analysis.
Where
more
than
one
quarter
of
a
cow
was
infected
the
quarters
were
as-
sumed
to
be
independent.
This
is
because
many
of
the
infections
were
of
environmental
origin
and
infection
in
one
quarter
was
considered
unlikely
to
affect
the
probabil-
ity
that
another
quarter
would
be
infected.
Clinical
treatment
failure
was
defined
as
having
occurred
where
a
herd
owner
elected
to
re-treat
a
quarter
which
had
been
diagnosed
as
having
clinical
mastitis,
within
21
days
of
the
initial
diagnosis.
Cows
were
also
removed
from
the
trial
where
the
herd
owner
diagnosed
clinical
mastitis
in
another
quarter
within
21
days
of
initial
diagnosis.
A
fur-
ther
group
of
cows
were
removed
from
the
analysis
for
miscellaneous
reasons,
including
death
or
antibiotic
treat-
ment
for
other
conditions
(e.g.
retained
foetal
membranes,
lameness,
etc.).
Clinical
cure
rate
was
defined
as
the
in-
verse
of
the
proportion
of
clinical
quarters
that
were
re-
treated
by
the
herd
owners
within
21
days
of
diagnosis.
Failure
of
bacteriological
cure
was
defined
as
having
oc-
curred
where
the
same
bacterial
species
was
isolated
at
D14
or
D21
as
at
DO
in
a
quarter.
Quarters
from
which
the
initially
isolated
bacterial
species
was
not
re-isolated
at
D14
or
D21
were
further
divided
into
those
quarters
where
no
bacteria
at
all
were
isolated
at
D14
or
D21
(Cured
bac-ve
)
and
those
where
another
bacterial
species
was
isolated
at
either
D14
or
D21
(Cured
b.
e
)
.
Where
more
than
one
pathogen
species
was
isolated
a
quarter,
none
of
the
pathogens
were
to
be
present
at
D14
or
D21
for
a
cure
to
be
defined
as
having
occurred.
Quarters
defined
as
having
subclinical
mastitis
in
cows
in
which
clinical
quarters
were
treated
with
P/DHS
were
de-
fined
as
having
"nil"
treatment.
It
was
assumed
that
any
diffusion
of
the
P/DHS
from
the
clinical
quarter
to
the
other
quarter(s)
was
insignificant
and
would
not
have
altered
the
bacteriological
outcome
for
that
subclinical
quarter.
Categorical
data
(e.g.
cured/not
cured)
were
analysed
by
Chi-squared
statistics.
Continuous
data
(SCC
and
IQR)
were
analysed
by
repeated
measures
ANOVA
with
treat-
ment
and
clinical
status
(clinical
or
subclinical
mastitis)
as
the
main
effects.
This
experiment
was
designed
to
have
the
ability
(a
=
0.05,
=
0.8)
to
detect
a
15%
difference
in
bacterial
cure
rate
assuming
300
cows
per
treatment,
with
15%
of
cows
being
removed
from
analysis
because
no
bacteria
were
cultured
at
the
pre-treatment
sample
and
at
an
average
cure
rate
of
60%.
If
35%
of
all
pre-treatment
bacterial
isolates
were
Streptococ-
cus
spp.,
a
20%
difference
in
bacterial
cure
for
the
Streptococ-
cus
spp.
could
have
been
detected
(Aleong
and
Bartlett,
1979).
Results
A
total
of
595
cows
and
1480
quarters
were
enrolled,
in-
cluding
682
(46.0%)
subclinical
and
798
(54.0%)
clinical
quarters.
PEN
was
used
to
treat
301
cows
and
P/DHS
294
cows.
Each
cow
enrolled
had
an
average
of
2.49
quarters
identified
as
having
mastitis,
of
which
1.34
quarters
were
diagnosed
as
having
clinical
mastitis.
Over
85%
of
cows
had
more
than
one
quarter
with
mastitis
(clinical
and
sub-
clinical)
and
23.2%
had
more
than
one
quarter
with
clini-
cal
mastitis.
There
were
significantly
more
rear
(66.4%)
than
fore-quarters
(33.6%)
diagnosed
with
clinical
masti-
tis
(x
2
=
136.6;
p
<
0.001).
Of
the
total
of
812
quarters
from
which
bacteria
were
isolated,
92.5%,
7.0%
and
0.5%
had
one,
two
or
three
bacterial
species
isolated,
respec-
tively.
A
single
major
pathogen
was
isolated
from
69.9%
quarters,
two
major
pathogens
from
1.8%
quarters,
a
ma-
jor
and
minor
pathogen
from
5.4%
quarters,
a
single
mi-
nor
pathogen
from
22.6%
quarters
and
two
or
more
minor
pathogens
from
0.2%
quarters.
Ninety-eight
(16.5%)
cows
(248
quarters)
were
removed
from
the
trial
due
to
treatment
failure
in
the
clinical
quar-
ter,
mastitis
in
another
quarter
or
for
other
non-mastitis
related
reasons.
The
total
number
of
cows
removed
did
not
differ
(p
>
0.05)
between
the
PEN-treated
group
(n
=
52;
17.3%)
and
the
P/DHS
group
(n
=
46;
15.6%).
The
number
of
cows
removed
due
to
herd
owner
diagnosis
of
treatment
failure
did
not
differ
between
treatments
(13.3%
v.
8.8%
of
all
clinical
quarters
for
PEN
and
P/DHS,
re-
spectively;
p
>
0.05).
More
cows
treated
with
P/DHS
were
removed
due
to
treatment
of
another
quarter
than
PEN-treated
cows
(5.1%
v.
2.0%,
respectively,
p
<
0.05).
Following
removal
of
these
cows,
there
was
no
difference
between
treatment
types
in
the
age,
breed
of
animals
or
days
after
calving
on
day
of
diagnosis
(p
>
0.4,
p
>
0.5,
p
>
0.1,
respectively).
Streptococcus
uberis
was
the
predominant
isolate
from
both
clinical
and
subclinical
quarters
(Table
I).
The
cure
rate
of
Table
I.
The
number
of
bacteria
isolated
and
%
of
each
species
isolated
within
each
treatment
group
from
quarters
diagnosed
with
subclinical
or
clinical
mastitis
before
starting
treatment
(DO)
with
nothing
(Nil),
penethamate
hydriodide
(PEN)
or
penicillin
and
dihydrospreptomycin
(P/DHS)
Treatment
Subclinical
Clinical
PEN
Nil
PEN
P/DHS
Number
of
"no
growths"
234
213
130
90
Total
number
of
bacteria
isolated
111
123
284
294
Streptococcus
uberis
(%)
55.0
41.5
72.2
77.6
Staphylococcus
aureus
(%)
-
penicillin-sensitive
3.6
1.6
3.5
1.4
Staphylococcus
aureus
(%)
-
penicillin-resistant
0.0 0.0
0.7
0.7
Streptococcus
dysgalactiae
(%)
1.8
1.6
3.5
4.1
Coliforms
(%)
2.7
1.6
6.3
3.1
Coagulase-negative
staphylococci
(%)
25.2
34.1
11.6
8.8
Othersa
(%)
11.7
19.6
2.2
4.4
a
Including
Pseudomonas
aeruginosa,
Actinobacillus
spp.,
Bacillus
spp.
and
Corynebacterium
spp.
1998
New
Zealand
Veterinary
Journal
229
clinical
SU
infections
(p
>
0.2;
Table
II)
or
all
major
Gram-
positive
pathogens
(p
>
0.2)
did
not
differ
between
the
treat-
ments.
P/DHS
cured
a
greater
percentage
of
clinical
CNS
cases
than
did
PEN
(Table
II)
which
resulted
in
higher
over-
all
bacteriological
cure
rate
for
P/DHS
than
for
PEN
(84.9%
v.
76.4%,
x
2
=
3.92,
p
<
0.05;
Table
II).
A
higher
proportion
of
PEN
than
P/DHS
treated
quar-
ters
were
cured
but
another
pathogen
was
cultured
from
that
quarter
at
D14
and/or
D21
(i.e.
the
percentage
of
cured
quarters
that
were
cure
bac+ve
was
27.5%
v.
15.9%,
respec-
tively,
x
2
=
5.5;
p
<
0.05).
PEN
cured
more
cases
(77.3%
v.
63.8%;
p
<
0.05)
of
sub-clinical
infection
than
did
the
"nil"
treatment
(i.e.
the
subclinical
quarters
within
cows
where
the
clinical
quarter(s)
were
treated
with
P/DHS;
Table
II).
Resistance
to
penicillin
(16.6%
of
isolates)
and
strepto-
mycin
(13.3%
of
isolates)
occurred
in
some
SA
isolates.
One
SA
isolate
was
resistant
to
both
penicillin
and
strep-
tomycin.
Staphylococcus
aureus
was
isolated
from
14
herds,
and
three
of
these
herds
had
SA
isolates
with
two
differ-
ent
patterns
of
resistance.
All
SU
isolates
were
strepto-
mycin-resistant
and
were
sensitive
to
all
other
antibiotics
tested.
All
E.
coli
isolates
tested
were
penicillin-resist-
ant,
four
(25%)
had
amoxycillin
resistance,
and
one
(8.3%)
each
were
resistant
to
streptomycin
and
tetra-
cycline.
Somatic
cell
counts
Of
the
initial
1480
quarters
sampled
for
SCC
analysis,
413
(27.9%),
11
(1.7%)
and
five
(0.9%)
samples
on
DO,
D14
and
D21,
respectively,
were
unable
to
be
tested
due
to
excessive
numbers
of
clots
preventing
SCC
counting.
No
difference
in
SCC
on
DO,
D14
or
D21
was
found
between
cows
treated
with
PEN
or
P/DHS
(p
>
0.3),
ei-
ther
for
clinical
or
subclinical
quarters
(Figure
1a).
The
log
SCC
decreased
with
time
(p
<
0.05),
and
the
treat-
ments
declined
at
similar
rate
(p
>
0.2).
Quarters
diag-
nosed
as
having
subclinical
mastitis
had
significantly
lower
log
SCC,
than
clinical
quarters
at
DO,
D14
and
at
D21
(p
<
0.0001).
Conductivity
No
difference
in
the
IQR
was
found
for
clinical
or
subclini-
cal
quarters
treated
with
PEN
or
P/DHS
at
DO,
D14
or
D21
(p
>
0.2;
Figure
lb).
The
IQR
decreased
with
time
(p
<
0.05),
and
the
treatments
declined
at
similar
rate
(p
>
0.6).
Clinical
data
Milk
clots
were
detected
at
D
l4
and
D21
in
one
quarter
of
9
and
13
animals,
respectively,
and
there
was
no
difference
between
treatment
groups
in
the
proportion
of
quarters
with
clots
at
any
time
(p
>
0.5).
No
blood
in
the
milk
or
udder
oedema
were
detected
in
any
cow
at
D14
or
D21.
Table
II.
The
number
of
quarters
(n)
and
the
bacterial
cure
rate
(Cure
%)
following
treatment
of
quarters
with
subclinical
or
clinical
mastitis
with
either
nothing
(Nil),
penethamate
hydriodide
(PEN)
or
penicillin
and
dihydrospreptomycin
(P/DHS).
The
cure
rate
is
further
categorised
as
cure
with
no
bacteria
isolated
at
D14
and
D21
(Cured
bac
or
cured
with
new
infection
(Cured
bac
j
for
the
major
Gram-positive
isolates
and
for
isolates
combined
(All)
Bacteria
DO
Subclinical
Clinical
PEN
Nil
pa
PEN
P/DHS
pa
Streptococcus
uberis
n
46
41
0.29
126
151
0.6
Cure
(%)
82.6
73.2
81.7
84.1
Staphylococcus
aureus
n
2
1
NAb
7
4
0.2
Cure
(%)
100 100
14.3
50
Streptococcus
dysgalactiae
n
0
0
NA
2
6
0.35
Cure
(%)
50
83.3
Coliforms
n
3
1
NA
5
4
NA
Cure
(%)
100
50
100
100
CNS
n
22
31
0.15
15
12
0.03
Cure
(%)
81.8
64.5
53.3
91.7
Others°
n
12
19
0.33
2
8
NA
Cure
(%)
25
42.1
100
100
Major
Gram-positivesd
n
48
42
0.27
136
161
0.24
Cure
(%)
83.3
73.8
77.2
83.2
Cured
baci-ve
16.7
16.7
22.8
13
Cured
bac-ve
66.7
57.1
54.4
70.2
All
n
85
94
0.10
157
185
0.05
Cure
(%)
75.3
63.8
76.4
84.9
Cured
bac+ve
12.9
14.9
20.6
13.5
Cured
62.4
48.9
54.4
71.4
bec-ve
a
Probability
of
a
difference
in
cure
rate
among
treatments
within
subclinical
or
clinical
infected
quarters.
b
NA
indicates
that
there
is
0
value
and
hence
the
x2
statistic
could
not
be
calculated.
c
Including
Pseudomonas
aeruginosa,
Actinobacillus
spp.,
Bacillus
spp.
and
Cotynebacterium
spp.
d
SA,
SU
and
SD.
New
Zealand
Veterinary
Journal
1998
230
SCC
(x
1000)
ce
lls
/m
l
Discussion
The
hypothesis
that
no
difference
in
clinical
outcomes,
bacteriological
cure
rate
or
cytological
measures
for
peni-
cillin-sensitive
clinical
cases
of
clinical
mastitis
treated
with
PEN
or
P/DHS
occurred
was
accepted.
The
herd
owners
were
asked
to
re-treat
and
record
any
quarter
they
believed
had
failed
to
cure
on
the
basis
of
clinical
signs.
Using
continuation
of
clinical
symptoms
as
a
measure
of
clinical
failure
of
treatment,
there
was
no
difference
in
clinical
cure
rate
between
the
treatments.
Additionally,
there
was
no
difference
in
the
number
of
quar-
ters
detected
with
milk
changes
by
technicians
when
the
quarters
were
re-examined
at
D14
and
D21.
However,
more
of
the
quarters
with
subclinical
infections
that
were
left
untreated
(i.e.
the
subclinical
quarters
in
the
P/DHS-
treated
cows)
had
treatment
failures
than
sub-clinical
quar-
ters
from
cows
treated
with
PEN.
Streptococcus
uberis
was
the
most
commonly
isolated
patho-
gen
in
this
trial.
As
an
environmental
pathogen
it
can
sur-
vive
in
extra-mammary
sites,
including
cows'
skin,
the
res-
piratory
tract
and/or
faeces
(King,
1981;
Robinson
et
al.,
1985).
Pankey
etal.
(1996)
isolated
SU
or
SD
from
67.6%
of
clinical
cases
of
mastitis
in
heifers
within
5
days
of
calv-
ing,
which
is
in
close
agreement
with
the
present
trial.
In
contrast,
a
survey
performed
in
the
1960s
(Brookbanks,
1966)
isolated
only
22
SU
from
1137
bacterial
positive
quarters,
with
SA,
SD
and
Streptococcus
agalactiae
being
more
prevalent.
The
prevalence
of
bacterial
species
in-
volved
with
mastitis
appears
to
have
altered
in
the
last
30
years
in
New
Zealand
when
the
data
of
Brookbanks
(1966)
and
the
present
data
set
are
compared.
Introduction
of
mastitis
control
strategies
such
as
the
five
point
plan
and
the
SAMM
plan
(Woolford
et
al.,
1995)
have
altered
the
prevalence
of
mastitis
pathogens
in
New
Zealand.
Simi-
larly
mastitis
programmes
based
on
the
five
point
plan
have
resulted
in
changes
in
bacteria
species
causing
mastitis
and
resulted
in
significant,
economically
effective
reductions
in
the
prevalence
of
mastitis
internationally
(Benson
and
Dettman,
1977;
Morris
etal.,
1978;
Blowey,
1986).
The
cure
rates
for
SU
are
reported
to
be
72-77%
internation-
ally
(Craven,
1987;
King,
1981).
The
present
trial
achieved
a
cure
rate
of
greater
than
80%.
The
present
data
indicate
that
combination
of
an
another
antibiotic
(e.g.
dihydro-
14000
12000
10000
8000
6000
4000
2000
0
DO
D14
D21
streptomycin)
with
penicillin
did
not
increase
cure
rates
of
SU
above
that
achieved
by
penicillin
alone,
which
is
in
agreement
with
the
review
by
King
(1981).
The
second
most
common
pathogen
isolated
was
CNS,
in
agreement
with
a
previous
study
in
heifers
in
New
Zea-
land
(Pankey
et
al.,
1996).
Coagulase-negative
staphylo-
cocci
are
often
defined
as
"minor"
pathogens
(Sandholm
et
al.,
1980),
implying
that
they
result
in
little
cellular
dam-
age
or
immune
response
following
infection.
However,
sufficient
clinical
signs
were
present
for
herd
owners
to
detect
CNS
infection,
indicating
that
CNS
are
more
patho-
genic
than
previously
believed.
It
is
not
known
how
many
subclinical
cases
of
CNS
were
present
in
these
herds
and
hence
what
proportion
of
these
infections
produce
changes
in
the
udder
or
milk
that
are
clinically
evident.
Cure
rates
of
minor
pathogens
are
reported
to
be
in
the
range
of
51-
90%
(Craven,
1987;
Jarp
et
al.,
1989).
The
cure
rate
for
CNS
achieved
by
P/DHS
(91.7%)
was
higher
than
that
for
clinical
quarters
treated
with
PEN
(8/16,
50%)
and
high
compared
to
previous
studies.
Paradoxically
the
cure
rate
of
the
"nil"
treated
quarters
infected
with
CNS
was
74.5%
and
for
subclinical
quarters
treated
with
PEN,
81.8%.
Eight
species
of
CNS
have
been
identified
in
milk
from
mastitis
cows
in
New
Zealand
and
the
resistance
to
penicillin
among
these
species
of
isolates
is
reported
to
range
from
0%
to
75%
(Hodges
et
al.,
1984).
It
is
not
known
whether
CNS
species
vary
in
pathogenicity
in
New
Zealand
and
hence
whether
the
proportion
of
CNS
species
isolated
and
their
resistance
patterns
varies
between
sub-
clinically
and
clinically
affected
quarters.
The
overall
bacteriological
cure
rates
of
greater
than
75%
achieved
in
this
trial
are
comparable
to
those
reported
in-
ternationally
(Craven,
1987;
Faull
and
Ward,
1975;
Harris
et
al.,
1977;
Jarp
a
al.,
1989).
The
high
proportion
of
SU
and
minor
pathogens
is
likely
to
have
contributed
to
this
relatively
high
cure
rate.
The
cure
rate
for
SA
infections
treated
during
lactation
are
reported
to
be
between
25%
and
35%
(Craven,
1987),
thus
if
a
high
proportion
of
clinical
cases
had
been
due
to
SA,
the
overall
cure
rate
would
prob-
ably
have
been
lower.
PEN
was
equally
effective
as
P/DHS
in
inducing
bacte-
riological
cure
of
infections
due
to
SU
and
due
to
major
Gram-positive
(SU,
SA
or
SD)
pathogens,
but
less
effec-
tive
in
curing
CNS
infections.
This
resulted
in
a
lower
100
-
80
-
60
-
40
-
20
-
DO
D14
D21
IQ
R
(
%
inc
rea
se)
Figures
la
and
lb.
The
geometric
mean
and
upper
95%
confidence
interval
for
the
somatic
cell
count
(left)
and
interquarter
conductivity
(right)
for
clinical
quarters
treated
with
penethamate
hydriodide
(
)
or
penecillin
and
dihydrospreptomycin
(
)
and
subclinical
quarters
treated
with
penethamate
hydriodide
(
)
or
nothing
(
).
1998
New
Zealand
Veterinary
Journal
231
overall
cure
rate
for
PEN-treated
quarters.
The
reasons
for
this
difference
in
bacteriological
cure
rate
are
not
clear.
The
inclusion
of
dihydrostreptomycin
in
P/DHS
provides
a
wider
spectrum
of
activity
than
PEN.
For
example,
coliforms
are
routinely
resistant
to
penicillin,
yet
only
one
of
12
isolates
tested
were
streptomycin-resistant.
The
treatments
may
have
differed
in
the
period
over
which
new
infections
were
prevented
from
being
established
or
over
which
existing
bacterial
infection
could
not
be
cultured.
For
example,
dihydrostreptomycin
may
be
detectable
for
up
to
7
days
in
milk
following
intramammary
infusion
of
200
mg
in
oil
(Mercer
et
al.,
1976),
significantly
shorter
than
the
2
days
following
PEN
treatment.
More
of
the
PEN-treated
cows
that
did
cure
(i.e.
the
bacterial
species
isolated
on
DO
was
not
re-isolated
on
D14
or
D21)
were
infected
with
a
differ-
ent
bacterial
species
on
D14
or
D21
than
P/DHS-treated
cows.
Similarly,
more
quarters
treated
with
PEN
may
have
been
re-infected
with
the
same
bacterial
species
than
quar-
ters
treated
with
P/DHS.
If
bacterial
cure
did
occur,
but
re-
infection
with
the
same
bacterial
species
occurred
before
D21,
then
this
could
not
be
distinguished
from
a
"failure
to
cure"
and
would
be
classified
as
such.
Route
of
administration
may
also
have
affected
the
new
infection
rate
or
re-infection
rate.
Treatment
with
intramammary
preparations
may
remove
more
bacteria
from
the
teat
canal
than
parenteral
treatment.
Removal
of
bacteria
from
the
teat
canal
may
reduce
the
incidence
of
new
infections
or
re-infections,
as
most
infections
com-
mence
with
colonisation
of
the
teat
canal
(Lacy-Hulbert
and
Hillerton,
1995).
Subclinical
quarters
in
cows
which
were
treated
with
PEN
had
a
higher
cure
rate
in
the
subclinical
quarters
from
cows
treated
with
P/DHS
(i.e.
77.3%
v.
63.8%).
This
indicates
an
advantage
of
a
parenteral
treatment
for
clinical
mastitis
which
incidentally
will
lead
to
higher
cure
rates
in
sub-
clinical
quarters
which
either
may
not
be
detected
at
all,
or
if
detected,
not
routinely
treated.
The
actual
financial
ben-
efit
of
using
a
parenteral
treatment
compared
to
intramammary
treatment
will
depend
on
the
number
of
quarters
infected
either
subclinically
or
clinically,
the
rela-
tive
costs
of
the
treatments,
the
relative
cure
rates
and
the
cost
of
antibiotic
containing
milk
discarded.
Penicillin
resistance
was
not
detected
in
the
small
number
of
SU
isolates
tested.
Penicillin
resistance
has
been
de-
tected
in
0.5%
of
a
large
number
of
SU
isolates
from
clini-
cal
mastitis
submitted
to
a
local
laboratory
(R.
Ellison,
pers.
comm.).
Although
SU
is
inherently
resistant
to
strep-
tomycin,
this
is
not
of
major
concern,
as
no
streptomycin-
only
intramammary
preparations
are
available
in
New
Zea-
land.
The
resistance
patterns
of
the
30
SA
isolates
tested
were
variable
within
and
between
herds.
Variation
in
cure
rate
between
cows
and
between
herds
may
be
at
least
partly
due
to
variation
in
pattern
of
SA
resistance.
However,
SA
represented
only
2.9%
of
the
bacterial
positive
cases
of
mastitis
or
1.7%
of
all
presented
cases
of
clinical
mastitis.
About
16.7%
of
the
SA
isolates
were
resistant
to
penicillin
and
about
38%
of
New
Zealand
CNS
isolates
are
reported
to
be
penicillin-resistant
(Hodges
et
al.,
1984).
All
E.
coli
tested
were
penicillin-resistant,
as
expected.
The
prevalence
of
clinical
mastitis
cases
with
penicillin-resistance
may
be
estimated
as
the
sum
of
all
the
coliforms
(i.e.
3.9%),
a
quarter
of
the
SA
isolates
(0.17
x
3.7%)
and
38%
of
the
CNS
isolates
(0.38
x
8.1%)
producing
a
total
of
7.6%
of
the
bacteria
positive
or
5.1%
of
all
cases
of
clinical
masti-
tis
cases.
Penicillin
resistance
was
present
in
only
a
small
percentage
of
the
total
population
of
clinical
cases.
There
was
no
difference
between
treatments
in
the
indi-
vidual
quarter
SCCs
or
in
conductivity
at
DO,
D14
or
D21.
This
indicates
that
the
two
treatments
were
equally
effec-
tive
in
reducing
the
antigenic
stimulus
(i.e.
bacteria)
that
led
to
elevations
in
SCC
and
in
allowing
repair
of
the
udder
and
hence
a
reduction
in
conductivity
(Lacy-Hulbert
and
Woolford,
1997).
The
SCC
is
especially
important
as
it
is
the
SCC,
rather
than
bacterial
infection
of
the
udder
as
such,
that
is
one
of
the
milk
quality
measures
used
rou-
tinely
by
the
milk
processing
industry.
In
conclusion,
treatment
of
clinical
and
subclinical
cases
of
mastitis
during
lactation
with
either
P/DHS
or
PEN
resulted
in
a
bacterial
cure
rate
of
greater
than
75%
and
a
clinical
cure
rate
of
greater
than
85%.
It
is
concluded
that
there
was
no
difference
in
cure
rate
between
P/DHS
or
PEN
for
penicillin-sensitive
clinical
cases
of
mastitis
in
early
spring
in
New
Zealand.
The
relatively
low
preva-
lence
of
penicillin-resistant
pathogens
suggests
that
peni-
cillin-based
antibiotics
can
be
used
to
treat
cases
of
clini-
cal
mastitis
in
the
Waikato
region
of
New
Zealand.
Acknowledgments
This
trial
was
conducted
with
the
approval
of
the
Animal
Ethics
Committee
of
AgResearch,
Ruakura.
Thanks
to
the
herd
owners
who
took
the
time
to
present
cows
for
examination
at
a
very
busy
time
of
year.
The
tech-
nical
assistance
of
Rhonda
Cooper,
Sabrina
Graham,
Julie
Vercoe
and
Annie
Watts
is
gratefully
acknowledged.
The
bacteriology
and
sensitivity
testing
was
performed
by
Al-
pha
Scientific,
Ellis
St,
Hamilton.
This
study
was
com-
pleted
with
the
significant
financial
assistance
of
Boehringer
Ingelheim
NZ
Ltd
and
the
role
of
Tom
Davies
and
Mike
Froger
from
BINZL
in
supporting
and
encouraging
the
trial
is
acknowledged.
Woody
Pankey
and
Jane
Lacy-Hulbert
had
many
constructive
comments
to
make
on
the
draft
manuscript.
References
Aleong
J,
Bartlett
DE.
Improved
graphs
for
calculating
sample
sizes
when
comparing
two
independent
binomial
distributions.
Bio-
metrics
35,
875-81,
1979.
Anon.
Guidance
document
for
target
animal
safety
and
drug
effective-
ness
studies
for
anti-microbial
bovine
mastitis
products
(lactat-
ing
and
non-lactating
cow
products).
Guidelines
49.
Office
of
New
Drug
Evaluation,
Center
for
Veterinary
Medicine,
Food
and
Drug
Administration,
Rockville,
1996.
Benson
RJ,
Dettmann.
The
New
South
Wales
mastitis
control
pro-
gram.
3.
The
effect
on
production
and
economics.
Australian
Veterinary
Journal
53,
538-41,
1977.
Blowey
RW.
An
assessment
of
the
economic benefits
of
a
mastitis
control
scheme.
Veterinary
Record
119,
551-3,
1986.
Brookbanks
E0.
A
report
on
surveys
of
the
incidence
of
mastitis
infec-
tion
in
New
Zealand
dairy
herds.
New
Zealand
Veterinary
Journal
14,
62-70,
1966.
Craven
N.
Efficacy
and
financial
value
of
antibiotic
treatment
of
bo-
vine
clinical
mastitis
during
lactation:
A
review.
British
Veteri-
nary
Journal
143,
410-22,
1987.
Faull
WB,
Ward
WR.
Treatment
of
clinical
mastitis:
Two
intramammary
formulations
compared.
Veterinary
Record
96,
127-9,
1975.
Harris
AM,
Davies
AM,
Marshall
MJ,
Evans
JM,
Hendy
PG,
Watson
DJ.
The
treatment
of
clinical
mastitis
with
cephaxazole
and
peni-
cillin.
The
Veterinary
Record
101,
4-7,
1977.
Hodges
RT,
Jones
YS,
Holland
JTS.
Characterisation
of
Staphyloco-
cci
associated
with
clinical
and
subclinical
bovine
mastitis.
New
Zealand
Veterinary
Journal
32,
141-5, 1984.
Jarp
J,
Bugge
HP,
Larsen
S.
Clinical
trial
of
three
therapeutic
regi-
mens
for
bovine
mastitis.
The
Veterinary
Record
124,
630-4,
1989.
232
New
Zealand
Veterinary
Journal
1998
King
JS.
Streptococcus
uberis:
A
review
of
its
role
as
a
causative
organ-
ism
of
bovine
mastitis.
2.
Control
of
infection.
British
Veterinary
Journal
137,
160-5,
1981.
Lacy-Hulbert
SJ,
Woolford
MW.
Mastitis
diagnosis
and
resolution.
In:
Hayes
DP,
McDougall
S
(eds).
Proceedings
of
the
Second
Inter-
national
Conference
for
the
Society
of
Dairy
Cattle
Veterinarians
of
the
New
Zealand
Veterinary
Association,
24-36,
1997.
Lacy-Hulbert
SJ,
Hillerton
JE.
Effect
of
physical
properties
of
the
teat
canal
on
susceptibility
to
intramammary
infection.
In:
Saran
A,
Soback
S
(eds).
Proceedings
of
the
Third
IDE
International
Mastitis
Seminar
Sl,
19-24, 1995.
Mercer
HD,
Geleta
JN,
Baldwin
RA,
Shimoda
W.
Viewpoints
and
current
concepts
regarding
accepted and
tried
products
for
con-
trol
of
bovine
mastitis.
Journal
of
the
American
Veterinary
Medi-
cine
Association
169,
1104-14,
1976.
Morris
RS,
Blood
DC,
Williamson
NB,
Cannon
CM,
Cannon
RM.
A
health
program
for
commercial
dairy
herds.
4.
Changes
in
mastitis
prevalence.
Australian
Veterinary
Journal
54,
247-51,
1978.
Murdough
PA,
Deitz
KE,
Pankey
JW.
Effects
of
freezing
on
the
viabil-
ity
of
nine
pathogens
from
quarters
with
subclinical
mastitis.
Journal
of
Dairy
Science
79,
334-6,
1996.
Pankey
JW,
Barker
RM,
Twomey
A,
Duirs
G.
Comparative
efficacy
of
dry-cow
treatment
regimens
against
Staphylococcus
aureus.
New
Zealand
Veterinary
Journal
30,
13-5,
1982.
Pankey
JW,
Pankey
PB,
Barker
RM,
Williamson
JH,
Woolford
MW.
The
prevalence
of
mastitis
in
primiparous
heifers
in
11
Waikato
dairy
herds.
New
Zealand
Veterinary
Journal
44(2),
41-4,
1996.
Robinson
TC,
Jackson
ER,
Marr
A.
Factors
involved
in
the
epidemi-
ology
and
control
of
Streptococcus
uberis
and
coliform
mastitis.
British
Veterinary
Journal
141,
635-42,
1985.
Sandholm
M,
Kaartinen
L,
Pyorala
S.
Bovine
mastitis
Why
does
antibiotic
therapy
not
always
work?
An
overview.
Journal
of
Vet-
erinary
Pharmacology
and
Therapeutics
13,
248-60,
1990.
Whittem
T,
Gaon
D.
Principles
of
antimicrobial
therapy.
Veterinary
Clinics
of
North
America:
Small
Animal
Practice
28,
197-213,
1998.
Williamson
JH,
Woolford
MW,
Day
AM.
The
prophylactic
effect
of
a
dry-cow
antibiotic
against
Streptococcus
uberis.
New
Zealand
Vet-
erinary
Journal
43,
228-34,
1995.
Woolford
MW.
Mastitis
in
New
Zealand.
Where
are
we
at
and
where
are
we
going?
In:
Hayes
DP,
McDougall
S
(eds).
Proceedings
of
the
Second
International
Conference
for
the
Society
of
Dairy
Cattle
Veterinarians
of
the
New
Zealand
Veterinary
Association,
1-14,
1997.
Woolford
MW,
Hook
IS,
Eden
MT,
Joe
AK.
The
"SAMM
PLAN"
a
seasonal
approach
to
managing
mastitis.
In:
Saran
A,
Soback
S
(eds).
Proceedings
of
the
Third
IDE
International
Mastitis
Semi-
nar
S4,
59-63,
1995.
Ziv
G
.
Practical
pharmacokinetic
aspects
of
mastitis
therapy.
II.
Prac-
tical
and
therapeutic
applications.
Veterinary
Medicine
and
Small
Animal
Clinician
75,
469-74,
1980.
Accepted
for
publication
20
November
1998.