160 / 232
seven (27%) of these patients had progression in at least
one ear, and 2 patients had progression in both ears.
Twenty-three of the 100 (23%) patients had ipsilateral
progression, whereas 6 of 100 (6%) had contralateral pro-
gression. As compared to patients with EVA and
ipsilateral hearing loss (65 of 164 [39.6%]), patients with
unilateral hearing loss without EVA had a significantly
lower rate of progression (27 of 100 [27%];
P
¼
.037).
There was no difference in the rate of progression in
patients with normal hearing and EVA (3 of 37 [8.1%])
and patients with unilateral hearing loss without EVA (6
of 100 [6%];
P
¼
.9).
Low-Frequency Hearing Loss Progression
Of the 144 patients with EVA, 237 ears had pure
tone audiometric data at 250 Hz (PT250); 194 of these
ears had hearing loss at 250 Hz. PT250 strongly corre-
lated with midpoint and operculum measurements
(Spearman rho
¼
0.43,
P
<
.001 and 0.42,
P
<
.0001,
respectively). The PT250 also strongly correlated with the
final PTA and HFPTA (Spearman rho
¼
0.79,
P
<
.0001
and 0.64;
P
<
.0001, respectively). Controlling for
temporal bone measurements, the PT250 was still
highly correlated with the final PTA (Spearman rho
¼
0.7;
P
<
.0001). There was no association between the
PT250 and a positive
SLC26A4
test result.
One hundred seventy-six patients met our study
criteria for determining hearing loss progression. In these
patients, 57 ears showed progression (median, 40 dB
hearing loss at 250 Hz), and 119 ears showed no progres-
sion (median, 25 dB hearing loss at 250 Hz;
P
¼
.003). A
mixed hearing loss at 250 Hz was seen in 117 ears. There
was a significantly higher rate of progression in ears with
a hearing loss at 250 Hz than in ears with normal hearing
thresholds (46 of 117 [39.3%] vs. 11 of 59 [18.6%], respec-
tively;
P
¼
.0003).
Among all ears that had PT250 values and were an-
alyzed for progression (n
¼
176 ears) using GEE to
account for clustered data by ear, the odds for progres-
sion increased with increasing PT250 (Table IV). This
relationship weakened when midpoint measurements
and initial PTA were controlled for in the analysis. Both
the hearing level at 250 Hz and the initial PTA were
strongly correlated with the likelihood of progression
(Table IV).
Genetics
A summary of genetic test results is shown in
Tables V and VI.
SLC26A4
testing was positive for a sig-
nificantly higher number of patients with bilateral EVA
than with unilateral EVA (43.5% vs. 10.3%;
P
¼
.007). Of
patients with positive
SLC26A4
results, 7 of 24 (29%)
had biallelic mutations (i.e., Pendred syndrome), and 17
of 24 (71%) had a single mutation. No unilateral EVA
patients had biallelic mutations compatible with Pen-
dred syndrome. Very few patients had mutations in the
GJB2
,
GJB6
, or
MTRNR1
genes. Only 1 patient had
biallelic
GJB2
mutations.
Overall, the rate of hearing loss progression in
patients with
SLC26A4
mutations was similar to the
rate in patients without
SLC26A4
mutations (Table VII).
Analysis of hearing loss phenotypes indicated that the
rate of progression was significantly related to positive
results in ears with hearing loss in patients with bilat-
eral EVA (Table VIII). Additionally, of patients with
positive genetic test results and progressive hearing
loss, significantly more had bilateral EVA (14 of 35
[40%]) than unilateral EVA (3 of 30 [10%];
P
¼
.006).
DISCUSSION
Although unilateral EVA is not an uncommon
otologic finding, its audiometric and temporal bone
TABLE IV.
Risk Analysis of Hearing Loss Progression.
Variable
Odds Ratio
95% CI
P
PTT @ 250 Hz
1.10*
1.03-1.19
.009
PTT @ 250 Hz
†
1.08*
0.996-1.17
.06
PTT @ 250 Hz
‡
1.04*
0.93-1.16
.5
Initial PTA4
1.12*
1.04-1.20
.003
Initial PTA4
§
1.10*
1.02-1.19
.018
*For every 5-dB increase in initial value.
†
Controlling for midpoint.
‡
Controlling for initial PTA4 value.
§
Controlling for midpoint. CI
¼
confidence interval; PTA
¼
pure tone
average; PTT
¼
pure tone threshold.
TABLE V.
Genetic Test Results.
Unilateral
EVA, n
¼
74
Bilateral
EVA, n
¼
70
P
Number of patients who
received genetic testing
GJB2
51 (68.9%)
50 (71.4%)
.74
GJB6
4 (5.4%)
5 (7.1%)
.74*
SLC26A4
39 (52.7%)
46 (65.7%)
.11
MTRNR1
4 (5.4%)
3 (4.3%)
1.0*
Positive results among
those tested
GJB2
3 (5.9%)
3 (6.0%)
1.0*
GJB6
0
1 (20%)
1.0*
SLC26A4
4 (10.3%)
20 (43.5%)
0.0007
MTRNR1
0
0
—
*Fisher exact test.
EVA
¼
enlarged vestibular aqueduct.
TABLE VI.
SLC26A4
Genotypes.
3 –2 A
>
G
682–698del17
L445W
M1T
N322D (2)
1614
þ
1 G
>
A
P10T (2)
N324Y
Y530H
L50R (2)
F335L
L597S
G209V
F335S
L729P
T410M
1001
þ
1 G
>
A (2)
G740S
V138F
T416P (3)
G740V
L236P (2)
L441P
Amino acid changes shown for missense mutations. Numbers in
parentheses denote the number of alleles found.
Laryngoscope 123: June 2013
Greinwald et al.: Unilateral Enlarged Vestibular Aqueduct
138