240 / 280
abscesses
spanning
multiple
locations.
Confidence
intervals
around
the
point
estimates
are wide,
particularly
for
specificity,
due
to
the
small
sample
sizes
of
individual
locations.
No
formal
comparisons
of
test
performance
were
made
between
CT
and
ultrasound by
location
given
the
small
subgroup
sizes.
In
general,
the site-specific estimates are consistent with the overall estimates
in
which
sensitivity
is
similar
but
somewhat
lower
for
the
ultrasound method
compared
to
CT.
4. Discussion
This
study
shows
that ultrasound may be as
sensitive, yet more
specific,
than
CT
in
the
diagnosis
of
lateral
neck
abscesses when
compared
to
the
gold
standard,
drainage
of
the
abscess.
As
such,
practice
guidelines may be developed based upon
the
cost,
safety,
and
discomfort
of
the
two
procedures.
In
2010
in Oklahoma
City,
ultrasound
cost
$79.97
to
administer,
while
CT
administration
with contrast cost $220.11
[7]
.
It
is often necessary
to sedate a child
to
undergo
a
CT
scan,
adding
to
its
cost
and
associated
risk.
Contrast associated allergy, although
rare,
is a potential
side effect
of CT
[8]
. For children,
there may be an
increased
fear of CT because
they
have
to
be
separated
from
their
parent
or
guardian
for
an
extended
period
of
time.
Separation
anxiety
is
avoided
when
ultrasound
is
used.
There are many concerns about
the negative
long-term effects
of
radiation
from CT. Computed
tomography-related
x-ray doses
are
large
enough
that
there
is
statistically-significant
epidemio-
logical evidence of a
small
increase
in
lifetime attributable
risk of
cancer
incidence,
ranging
from 0.02%
in 80 year old men
to nearly
1%
in 20 year old women undergoing CT
[8]
. On average,
risks are
0.07%
larger
for
children
than
adults.
Annually,
out
of
600,000
children
in
the United States who
receive a head or abdominal CT,
it
is estimated
that 500 will die of cancer which
is directly
related
to
the CT
[9,10]
.
The
cumulative
radiation
exposure
from
two
to
three
head
CT
scans
in
children
under
15 may
triple
the
risk
of
brain
cancer
[11]
.
Ultrasound
avoids
the
risks
of
radiation.
One
drawback
to
ultrasound
is
that
a
probe must
be
placed
on
the
child’s neck;
this
could
cause
pain
or
discomfort
at
the
infection
site.
In
the
cases
examined
for
this
study,
the
performance
of
at
least
one
ultrasound
and
one
CT
was
limited
by
patient
movement.
When
examining
the
risks
and
benefits
of
both
techniques,
it
is
likely
that ultrasound may be preferred over CT
in
many
instances
for
the
diagnosis
of
pediatric
lateral
neck
abscesses.
We
demonstrate
in
this
review
that
ultrasound
may
have
greater
specificity when
compared
to CT
in
the diagnosis of
lateral
neck abscesses
in
children. This
is of great
importance clinically as
our
goal
for
imaging
is
often
to
determine who
does
not
need
to
undergo
surgical drainage.
In our population of children who were
already
treated
with
24 h
of
intravenous
clindamycin,
the
prevalence
of
abscesses
in
those
ultimately
requiring
incision
and drainage was 89%. Considering such a high prevalence,
reliably
finding
those children who do not have an abscess and are unlikely
to benefit
from
surgical drainage
is
critical. Although our numbers
for
specificity were
small
for
both
ultrasound
and
CT,
ultrasound
was
superior.
A
diagnostic
protocol
that
promotes
judicious
and
individual-
ized
use
of
ultrasound
and
CT
in
the
diagnosis
of
neck
abscesses
would
likely prove
to be beneficial
for
these
children. To decrease
cost,
discomfort,
and
potential
harm
to
the
child,
an
ultrasound
may
be
preferred
as
the
first
line
imaging
technique
in
many
situations.
Computed
tomography
may
be
useful
in
some
Table
3
Sensitivity
and
specificity
of
ultrasound
and
computed
tomography
by
abscess
location.
a
Abscess
location
(total)
Method
Sensitivity
Specificity
Counts
positive/total
b
Estimate
95%
CI
Counts
negative/total
c
Estimate
95%
CI
Anterior
cervical
(32)
Ultrasound 8/14
0.57
0.29–0.82 0/0
Not
estimable
CT
scan
9/16
0.56
0.30–0.80 0/2
0
0–0.84
Posterior
cervical
(21)
Ultrasound 1/3
0.33
0.008–0.91 1/1
1.0
0.025–1.0
CT
scan
11/16
0.69
0.41–0.89 0/1
0
0–0.98
Submandibular
and
submental
(30) Ultrasound 7/12
0.58
0.28–0.85 1/1
1.0
0.025–1.0
CT
scan
12/16
0.75
0.48–0.93 0/1
0
0–0.98
Parapharyngeal
(37)
Ultrasound 2/3
0.66
0.094–0.99 1/1
1.0
0.025–1.0
CT
scan
21/27
0.78
0.58–0.91 2/6
0.33
0.04–0.78
Parotid
(8)
Ultrasound 0/0
Not
estimable
0/0
Not
estimable
CT
scan
6/8
0.75
0.35–0.97 0/0
Not
estimable
Legend:
CI
–
confidence
interval;
CT
–
computed
tomography.
a
Twelve
imaging
studies
included
abscesses
from multiple
lateral
neck
locations
and
have
been
excluded
from
this
subgroup
analysis.
b
Sensitivity
data
presented
as
the
number
of
positive
tests
out
of
the
total
number with
a
gold
standard
positive
status.
c
Specificity
data
presented
as
the
number
of
negative
tests
out
of
the
total
number with
a
gold
standard
negative
status.
Table
2
Sensitivity,
specificity,
positive
and
negative
predictive
values
of
ultrasound
and
computed
tomography
as
compared
to
the
gold
standard,
drainage
of
abscess.
Method
Sensitivity
Specificity
Positive predictive
value
a
Negative
predictive
value
a
Counts
positive/total
b
Estimate 95%
CI
Counts
negative/total
c
Estimate 95%
CI
Estimate 95%
CI
Estimate 95%
CI
Ultrasound 18/34
0.53
0.35–0.70 5/5
1.0
0.48–1.00 0.96
0.86–0.99 0.16
0.10–0.23
CT
scan
61/90
0.68
0.57–0.77 2/11
0.18
0.02–0.52 0.88
0.85–0.91 0.06
0.02–0.19
Legend:
CI
–
confidence
interval;
CT
–
computed
tomography.
a
Assuming
abscess
prevalence
of
0.90.
b
Sensitivity
data
presented
as
the
number
of
positive
tests
out
of
the
total
number with
a
gold
standard
positive
status.
c
Specificity
data
presented
as
the
number
of
negative
tests
out
of
the
total
number with
a
gold
standard
negative
status.
B.
Collins
et
al.
/
International
Journal
of
Pediatric Otorhinolaryngology
78
(2014)
423–426
218