78 / 264
Endolymphatic sac decompression
was defined as complete
mastoidectomy with wide exposure of the posterior fossa dura
in the area of the endolymphatic sac but no incisions into the sac.
Shunting procedures
included mastoid shunt (opening the sac)
with or without silastic. If the author reported a sac decom-
pression procedure but described opening the endolymphatic
sac in the surgical technique, the study was included in the
mastoid shunting without silastic group.
If the surgical technique was unclear, we emailed the corre-
sponding author to ensure correct categorization of data.
Statistical Analysis
A meta-analysis of proportions was performed using MedCalc
12.7.0. Short-term and long-term endpoints were evaluated for the
following treatment groups:
1) Current ESS procedures
a. sac decompression versus mastoid shunt procedures
2) Current mastoid shunt procedures
b. mastoid shunt
with
silastic sheeting versus mastoid shunt
without
silastic sheeting
Each technique was weighted according to the number of pa-
tients treated. Analysis of pooled proportions was performed
where appropriate. MedCalc used a Freeman
Y
Tukey transfor-
mation (30) to calculate the weighted summary proportion under
the fixed and random effects model (31). Data were presented as
weighted proportions with corresponding 95% confidence in-
tervals (CIs). Both the fixed effects model and the random effects
model were used in this study.
In addition, a
W
2
test with Yates correction for continuity
was applied with 2-sided (or 2-tailed)
p
values for the com-
parison of 2 proportions from independent samples (these pro-
portions being the vertigo control and hearing preservation)
expressed as a percentage, as calculated from the aforemen-
tioned aggregations.
Complete methods are described in the Supplemental Methods
Section,
http://links.lww.com/MAO/A220.
RESULTS
The literature search resulted in a total of 4,262 man-
uscripts/abstracts. Of these, a total of 36 manuscripts
FIG. 6.
Vertigo control: Forest plot for mastoid shunt
with
silastic
with Category A/B vertigo control at a minimum of 12 months of
follow-up.
FIG. 7.
Hearing outcomes: Forest plot for mastoid shunt
with
silastic with hearing improved or stable at a minimum of 12 months
of follow-up.
FIG. 8.
Vertigo control: Forest plot for mastoid shunt
without
si-
lastic with Category A/B vertigo control at a minimum of 12 months
of follow-up.
FIG. 9.
Hearing outcomes: Forest plot for mastoid shunt
without
silastic with hearing improved or stable at a minimum of 12 months
of follow-up.
A. J. SOOD ET AL.
Otology & Neurotology, Vol. 35, No. 6, 2014
57