159 / 232
In the 144 patients with EVA, there were 243 ears
with hearing loss. The midpoint and operculum meas-
urements of these ears correlated with the PTA of the
final audiogram (Spearman rho
¼
0.20 [
P
¼
.002] and
0.17 [
P
¼
.007], respectively). This was consistent when
correlating midpoint and operculum measurements with
the HFPTA at the final audiogram (Spearman rho
¼
0.28 [
P
¼
.003] and 0.23 [
P
¼
.01], respectively).
When analyzing only ears with hearing loss and EVA
(n
¼
202), similarly significant correlations were found
for PTA measurements (Spearman rho
¼
0.18 [
P
¼
.01]
and 0.14 [
P
¼
.05], respectively). For HFPTA, the corre-
lations were slightly stronger (Spearman rho
¼
0.30
[
P
¼
.004] and 0.25 [
P
¼
.018], respectively). In the ears
of patients with EVA who had hearing loss at their ini-
tial evaluation (n
¼
164), significant correlations were
found between the PTA and the size of the midpoint and
operculum (Spearman rho
¼
0.22 [
P
¼
.006] and 0.26
[
P
¼
.0009], respectively). For HFPTA, similarly
significant correlations were found (Spearman rho
¼
0.32 [
P
¼
.005] and 0.24 [
P
¼
.04], respectively).
Vestibular aqueduct measurements were compared
among audiometric phenotypes (Table III). Only 1 patient
with unilateral EVA had contralateral hearing loss (mid-
point, 1.4; operculum, 2.0). The median midpoint
measurement in patients with ipsilateral hearing loss
(2.05; range, 0.1–4.1) was greater than this measurement
in patients with bilateral hearing loss (1.5; range, 0.1–
3.6;
P
<
.0001). Also, the median operculum measure-
ment in patients with ipsilateral hearing loss was greater
(2.45; range, 0.1–5.2) than this measurement in patients
with bilateral hearing loss (2.05; range, 0.4–7.5;
P
¼
.09);
however, this difference was not statistically significant.
Hearing Loss Progression
There were 232 ears in children with 3 months of
audiometric follow-up that were included in the analysis
of hearing loss progression; 31 ears were excluded from
analysis, as they had profound hearing loss (n
¼
201).
At the initial audiometric evaluation, 164 ears had hear-
ing loss, and 37 had normal hearing. Overall, 65 of 201
(39.6%) ears had progression. The initial median PTA in
the progressive hearing loss group was 53.3 (range,
15–90), and the final median PTA was 78.7 (range, 27.5–
120). The proportion of ears with progressive hearing
loss was slightly higher among ears of patients with
bilateral EVA compared to ears of patients with unilat-
eral EVA, although this difference was not statistically
significant (41 of 89 ears [46.1%] vs. 25 of 75 ears
[33.3%], respectively;
P
¼
.1). No difference in hearing
loss progression was found when we compared patients
with bilateral vs. unilateral hearing loss (55 of 134
[41%] vs. 11 of 30 [36.7%], respectively;
P
¼
.7).
Additionally, in analyzing patients with unilateral
hearing loss, we found a trend toward a higher preva-
lence of progression in patients with bilateral vs.
unilateral EVA, although this trend was not statisti-
cally significant (5 of 8 [62.5%] vs. 6 of 22 [27.3%],
respectively;
P
¼
.1). In the patients with unilateral
EVA and normal hearing, 3 of 37 (8.1%) demonstrated
hearing loss progression. However, the rate of progres-
sion in these patients was lower than in patients with
EVA and hearing loss at initial presentation (65 of 164
[39.6%];
P
¼
.0003) and in the ears of patients with
unilateral EVA (25 of 75 [33%];
P
¼
.002). Among the
75 ears in patients with unilateral EVA and hearing
loss, there was no difference in the likelihood of hear-
ing loss progression between ears with and without
EVA (16 of 48 [33.3%] vs. 9 of 27 [33.3%], respectively;
P
¼
1.0, Fisher exact test).
For all 201 ears analyzed for progression, the median
change in PTA between the initial and final audiogram
was 5.0 dB (range, 38.75 to 77.5 dB). For the ears that
progressed (n
¼
68), the annual rate of progression was
4.5 (range, 1.0–63 dB). For all 164 ears with initial hear-
ing loss, the median change in PTA between the initial
and final audiogram was 6.25 dB (range, 38.75 to 77.5);
among the ears with progression (n
¼
65), the annual
rate of hearing loss progression was 4 (range, 1.0–63 dB).
The rate of progression was significantly correlated with
the midpoint (Spearman rho
¼
0.41;
P
¼
.001), but not
with the operculum (Spearman rho
¼
0.18;
P
¼
.16). The
midpoint measurement was highly predictive of how fast
an individual would progress (taking into account the
clustering of ears or that two ears can belong to the same
individual;
b
¼
.37; standard error
¼
.07;
P
<
.0001). For
every 0.37-U increase in the midpoint measurement, the
rate of progression per year increased by a factor of 1 dB.
The change in PTA was not correlated with either the
midpoint (Spearman rho
¼
0.18;
P
¼
.16) or the opercu-
lum (Spearman rho
¼
0.08;
P
¼
.5).
During our study period, 100 patients with unilat-
eral SNHL without EVA who had at least 3 months of
follow-up audiometric data were identified. A portion of
these patients have been previously described.
32
Twenty-
TABLE III.
Median (Range) [IQR] Aqueduct Measurements by the Side of the Hearing Loss Relative to the Side of the EVA.
Hearing Loss
With EVA
Ipsilateral Opercular
Width, mm*
Ipsilateral Midpoint
Width, mm
†
Contralateral Opercular
Width, mm
Contralateral Midpoint
Width, mm
Ipsilateral hearing loss
2.7 (0.1–7.6) [2.1–3.5]
2.05 (0.1–4.1) [1.6–2.55]
1.25 (0.1–4.2) [0.85–1.85]
0.4 (0.1–3.8) [0.2–0.9]
Contralateral hearing loss
‡
1.1
0.7
2
1.4
Bilateral hearing loss
2.5 (0.4–7.5) [1.95–3.0]
1.5 (0.1–3.6) [1.0–1.9]
1.4 (0.1–1.9) [0.9–1.6]
0.4 (0.1–0.9) [0.1–0.7]
*Comparison between ipsilateral and bilateral hearing loss groups: not significantly different (
P
¼
.09).
†
Comparison between ipsilateral and bilateral hearing loss groups: significantly different (
P
<
.0001).
‡
Only 1 subject with unilateral EVA had contralateral hearing loss.
IQR
¼
interquartile range (25th and 75th percentile); EVA
¼
enlarged vestibular aqueduct.
Laryngoscope 123: June 2013
Greinwald et al.: Unilateral Enlarged Vestibular Aqueduct
137