Copyright © 2016 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
Letters to the Editor
Electrocochleography for the Diagnosis of Me´nie`re’s
Disease: The Wrong Stimulus
To the Editor: We wish to comment on the recent article
entitled ‘‘Click stimulus electrocochleography versus
MRI with intratympanic contrast in Menie`re’s disease:
A systematic review’’ by Ziylan et al. (1). The authors
discussed four relevant articles (2 – 5). The three employ-
ing click stimulus electrocochleography (EcochG) (2 – 4)
were used for their conclusion which was that ‘‘there is
relative low sensitivity and negative predictive value for,
click stimulus electrocochleography compared with MRI
with intratympanic gadolinium administration for detect-
ing endolymphatic hydrops in patients with Me´nie`re’s
disease.’’ It highlights the reasons for widespread scep-
ticism of electrocochleography as a reliable test for
Me´nie`re’s disease (6). Three authors (2–4) have used
a click SP/AP ratio of >0.33 which is not diagnostic for
endolymphatic hydrops and which will have enhanced its
apparent sensitivity.
The sensitivity and specificity figures for tone-burst
ECochG were not available from our data as we com-
pared the two tests with a clinical diagnosis of Me´nie`re’s
disease, rather than using magnetic resonance imaging
(MRI) as the ‘‘gold standard’’ of this condition. We have
subsequently revised our data in this manner to compare
tone-burst ECochG in our patients with an MRI scan
demonstrating endolymphatic hydrops. Table 3 in (1) has
been reproduced in Table 1 here to include our own
results for click and tone burst stimuli with a diagnostic
sensitivity for clicks of 28.6 and 92.9% for tone bursts. As
is mentioned in our article, we think that tone burst
ECochG may detect electrophysiologic changes from
pathologic basilar membranes with more sensitivity than
the degree of hydrops that can be detected currently with
MRI scanning.
That cochlear endolymphatic hydrops could be
ascertained by electrophysiological testing arose from
a publication by Gibson et al. (7) in 1977. Using trans-
tympanic electrocochleography they claimed that an
enlarged direct current component of the click stimulus
action potential (AP) called the summating potential
(SP) could be an indication of cochlear hydrops. This
was confirmed by Coats et al. (8) which lead to numerous
publications using an SP/AP ratio of 0.33 to 0.35, and to
an eventual disenchantment as to the reliability of the
test (6). It was further compounded by audiologists using
remote eardrum or canal electrodes that give responses
approximately one-tenth of the magnitude of those
achieved by direct contact with the cochlea. Sub-
sequently, Gibson (9) by comparing click response
SP/AP ratios of Me´nie`re’s ears with ears with compar-
able hearing instead of Me´nie`re’s patient opposite ears,
retracted this claim, unless the SP/AP ratio reaches a
severe criterion of 0.5.
Meanwhile, Gibson (9) had employed tone bursts as a
stimulus , giving the test a significantly higher sensitivity
and specificity, confirmed by others (5,10). It cannot yet
be achieved using a remote eardrum or canal electrode.
Nonetheless, click stimulus EcochG continues to be
employed as a diagnostic test for Me´nie`re’s disease
despite the evidence that it should abandoned.
The recent advent of MRI inner ear imaging to detect
hydrops in Me´nie`re’s disease (11) has predictably
inspired studies comparing its sensitivity with that of
electrocochleography (1). Images and conclusions from
MRI inner ear imaging seem to be confounded by
variables such as brand of scanner, head coil specifi-
cations, and the possibility that gadolinium entry may
be uneven and favor the vestibule. Meanwhile, trans-
tympanic tone burst electrocochleography remains
the most simple, cheap, and sensitive test for detecting
TABLE 1. Diagnostic values of electrocochleography compared with MRI with intratympanic gadolinium administration
Click Stimulus Sensitivity Specificity PPV NPV
Yamamoto et al. (2)
a
85.7 100% 100% 50.0%
Fukuoka et al. (3)
a
66.7% 100% 100% 25.0%
Seo et al. (4)
a
71.4% 80% 93.8% 40.0%
Hornibrook et al. (5)
b
28.6% 66.7% 44.4% 50.0%
Tone burst stimulus
Hornibrook et al. (5) 92.9% 26.7% 54.2% 80.0%
a
SP/AP diagnostic ratio >0.33.
b
SP/AP diagnostic ratio >0.5.
MRI indicates magnetic resonance imaging; NPV, negative predictive value; PPV, positive predictive value.
1677
Otology & Neurotology
37:1677–1681 ß 2016, Otology & Neurotology, Inc.