144 / 232
children
age
8–13
years
old.
When
limiting
the
multivariate
analysis
to
children
with
at
least
6
months
of
follow-up
the
difference
between
the
lowest
and
highest
age
group
was
significant
with
pre-school
children
having
an
incidence
of
perforation
more
than
5
times
higher
(95%
CI
1.68–17.93,
p
= 0.005)
than
that
of
children
aged
8–13
years
old.
The
perforation
rate
in
children
aged
5–7
years
old
was
not
significantly
different
than
that
of
children
aged
8–13
years
old
(
p
= 0.07).
Surgeon
acted
as
a
strong
negative
confounder
and
weakened
the
association
between
age
and
post-tympanoplasty
perforation,
thus
increasing
the
odds
of
perforation
in
pre-school
children
after
controlling
for
the
effect
of
surgeon.
5. Discussion
This
study
represents
the
largest
reported
evaluation
of
pediatric
tympanoplasty
assessing
specifically
the
outcome
of
pre-school
children.
After
adjusting
for
confounding
factors,
the
current
study
does
suggest
that
pre-school
children
have
a
significantly
higher
rate
of
post-tympanoplasty
perforation
as
compared
to older children. The higher
rate of
failure
in pre-school
children
seems
to
be
mostly
attributed
to
a
high
rate
of
re-
perforation noted more
than 6 months after
tympanoplasty, with a
third
of
failures
due
to
reperforations
after
initial
successful
healing.
This
could
be
explained
by
an
initially
successful
tympanoplasty
and
subsequent
reperforation
due
to
persistent
Eustachian
tube
dysfunction
or
due
to
an
acute
otitis media.
In
addition,
overall
success
rate
of
tympanoplasty
in
this
study was
lower
than
those
previously
reported
in
the
literature
for
all
age
groups. This may be partially explained by
the
significant
resident
involvement
in
those
cases
and
the
fact
that
the procedures were
performed
at
a
tertiary
referral
centre
and
that
some
of
these
children may have been
referred
from other otolaryngologists
that
may have considered
the
repair
too difficult
for
them
to attempt.
It
may
also
be
a
reflection
of
the
patient
selection
criteria
by
the
surgeons
at
our
centre.
Results
of
previous
studies
evaluating
the
association
between
age
and
success
rate
in
pediatric
tympanoplasty
have
been conflicting. A
study by Black et al which
included 14 children
age 2–7
years
old
revealed
a
rate
of
intact
graft
of 56%
in
that
age
group
as
compared
to
77%
in
children
age
8–10
years
old
[15]
.
The
difference
in
intact
graft
between
the
age
groups
was
not
statistically
significant.
Kessler
et
al.
examined
tympanoplasty
outcome
in
209
children
and
reported
a
lower
long-term
success
rate
in
the 37
children under 6 years old
(
p
<
0.05), but
the
short-
term
outcome
was
similar
between
the
different
age
groups
examined
[22]
. The
late
failure
rate observed
in
that
study and
the
current
study
suggest
a
high
rate
of
recurrent
perforation
after
initial
successful healing
possibly due
to
an
immature
Eustachian
tube
in
younger
children.
A
meta-analysis
evaluating
the
effect
of
age
on
pediatric
tympanoplasty success
included 19 articles evaluating
the effect of
age, with
the
lowest age
limit being 6 years old
[27]
. Compilation of
these
articles
revealed
a
linear
association
between
success
rate
and
increasing
age
(
p
= 0.005).
Interestingly,
only
5
out
of
the
30
articles
included
in
the
overall meta-analysis
had
reported
an
association
between
age
and
tympanoplasty
success
rate. While
the
literature
does
seem
to
indicate
that
the
success
rate
in
children
is
somewhat
lower
than
that
in
adults,
uncertainty
remains
as
to what
is
the
ideal
age
to
repair
tympanic membrane
perforations
in
children
[28]
.
The
current
results
suggest
that
surgeons
should
exert
caution
when
considering
performing
tympanoplasty
on
children
less
than
5
years
of
age
since
results
from
tympanoplasty
in
that age group
seems
to be associated with
a
lower
success
rate.
Possible
reasons
for
previous
failure
to
identify
an
association
between
age
and
pediatric
tympanoplasty
success
rate
include
exclusion
of
younger
children
(under
6
or
8
years
old),
lack
of
sufficient
power
to
detect
a
significant
difference as most studies published
thus
far had case series of
less
than
100
patients
and
lack
of multivariate
analysis
to
control
for
possible
confounding
factors.
While
complete
closure
of
the
tympanic
membrane
is
the
desired
goal,
significant
improvements
in
quality
of
life
can
also
be
achieved
with
a
decreased
perforation
size.
Sheahan
et
al.
conducted
a
phone
survey
with
parents
of
children
who
had
previously
undergone
a
tympanoplasty
to
evaluate
parental
satisfaction
8–60 months
after
surgery
[5]
.
Seventy-nine
percent
of
parents were
satisfied with
the
outcome.
For
children with
a
persistent perforation, 40% of parents were
satisfied, 56%
reported
fewer
ear
infections
and
40%
reported
improvement
in
hearing.
The
main
confounding
factor
that
modified
the
association
between age and perforation post-tympanoplasty was
the
identity
of
the
surgeon
performing
the
surgical
procedure.
The
fact
that
the
association
between
the
surgeon
and
the
anatomical
success
rate was maintained
in
the multivariate analysis
suggests
that
the
surgeon’s
effect
is
likely
attributable
to
criteria
used
for
patient
selection by each surgeon, surgical skills and variation
in
technique
undetected
by
the
chart
review
and
not
solely
due
to
patient’s
age or choice of graft material.
Individual surgeon’s success rate did
seem
to
indicate
that
surgeons
with
more
than
20
years
of
experience
(surgeons 1 and 2) had a higher
success
rate, which
is a
finding
that has previously been
reported when
comparing
senior
Table
3
Odds
ratio
of
various
complications
per
age
group.
Outcome
Age
2–4 OR
(95%CI)
Age
5–7 OR
(95% CI)
Age
8–13 OR
p
-Value
(overall)
Persistent
perforation
1.16
(0.49–2.73)
1.60
(0.85–3.02)
1.0
0.32
Recurrent
perforation
3.86
(0.88–16.88)
1.93
(0.49–7.67)
1.0
0.19
Any
perforation
1.66
(0.77–3.59)
1.74
(0.96–3.16)
1.0
0.16
Tubes
1.59
(0.48–5.29)
0.56
(0.17–1.82)
1.0
0.28
Cholesteatoma
2.27
(0.44–11.67)
3.46
(0.95–12.59)
1.0
0.12
Any
failure
1.58
(0.78–3.23)
1.72
(0.99–2.98)
1.0
0.13
Revision
surgery
2.01
(0.88–4.60)
1.47
(0.75–2.91)
1.0
0.24
Table
4
Pre-
and
post-operative
SRT
per
age
group.
Age
group
Pre-operative
SRT Post
operative
SRT
p
-Value
2–4
23.3
16.7
0.02
5–7
21.7
16.7
<
0.001
8–13
25.6
18.4
<
0.001
ANOVA
p
-value
0.23
0.61
Table
5
Pre-
and
post-operative
ABG
per
age
group.
Age
group
Pre-operative
ABG Post
operative
ABG
p
-Value
2–4
15.7
9.1
0.04
5–7
16.7
9.5
<
0.001
8–13
21.5
11.3
<
0.001
ANOVA
p
-value
0.16
0.24
M. Duval
et
al.
/
International
Journal
of
Pediatric Otorhinolaryngology
79
(2015)
336–341
122