Clinical Neurophysiology 123 (2012) 1872–1879

Contents lists available at SciVerse ScienceDirect

Clinical Neurophysiology
journal homepage: www.elsevier.com/locate/clinph

Ocular and cervical vestibular evoked myogenic potentials in multiple

sclerosis patients
Sibel Gazioglu ⇑, Cavit Boz
Department of Neurology, Karadeniz Technical University Medical Faculty, 61080 Trabzon, Turkey

See Editorial, pages 1693–1694

a r t i c l e i n f o h i g h l i g h t s

Article history:  Multiple sclerosis patients showed high frequency of abnormality in vestibular evoked myogenic poten-
Available online 12 March 2012 tials (VEMPs) tests, especially in ocular VEMP tests.
 VEMP abnormalities were not correlated with brainstem clinical or magnetic resonance imaging
Keywords: lesions.
Multiple sclerosis  VEMP abnormalities were significantly correlated with expanded disability status scale (EDSS).
Vestibular evoked myogenic potentials
(VEMPs)
Brainstem a b s t r a c t
Magnetic resonance imaging
Objective: Vestibular evoked myogenic potentials (VEMPs) are thought to provide useful information
about brainstem functions, as the neural pathways of both ocular and cervical VEMPs pass through the
brainstem. The aim of this study was to investigate the clinical value of ocular and cervical VEMP tests
in the evaluation of brainstem involvement in multiple sclerosis (MS) patients and to assess their relation
with clinical and cranial MRI findings.
Methods: Ocular and cervical VEMPs were recorded in 62 MS patients and 35 age and sex matched
healthy volunteers. The latencies, amplitude asymmetry ratios of both VEMP responses and abnormality
ratios (prolonged latencies and absent responses) were compared between the MS patients and the
control group and among the groups of MS patients.
Results: oVEMP mean n1 and p1 latencies and cVEMP mean p13 latency were significantly prolonged in
MS patients. Although the abnormality ratios of both VEMPs were higher in patients with brainstem clin-
ical or MRI lesions, the correlation was not statistically significant. Both ocular and cervical VEMP laten-
cies were significantly correlated with expanded disability status scale.
Conclusions: Although there is no significant correlation with clinical or MRI findings, MS patients show
high frequency of abnormality in VEMP tests, especially in oVEMP tests.
Significance: VEMP tests may be useful as an adjunct test in the evaluation of brainstem dysfunction in
MS patients.
Ó 2012 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights
reserved.

1. Introduction besides being an objective indicator of involvement of the related

pathways, evoked potentials are also important in the diagnosis
Multiple sclerosis (MS) is a chronic demyelinating disease of MS in terms of demonstrating subclinical demyelination
involving the white matter of the central nervous system. The signs (Chiappa, 1984; Comi et al., 1999; Fuhr and Kappos, 2001).
and symptoms of the disease, clinical course and response to treat- Vestibular evoked myogenic potentials (VEMPs) are short
ment are different in every case. Although magnetic resonance latency electromyographic (EMG) responses that can be recorded
imaging (MRI) is the most important test in the diagnosis of MS, from various muscles during the contraction phase in response to
acoustic stimulus (Debatisse et al., 2005; Rosengren et al., 2010;
⇑ Corresponding author. Tel.: +90 462 3775084; fax: +90 462 3775689. Akin et al., 2004; Zhou and Cox, 2004; Brantberg, 2009; Cherchi
E-mail address: [email protected] (S. Gazioglu). et al., 2009). VEMPs recorded from ipsilateral SCM muscle known

1388-2457/$36.00 Ó 2012 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.clinph.2012.01.022
 S. Gazioglu, C. Boz / Clinical Neurophysiology 123 (2012) 1872–1879 1873

as ‘‘cervical VEMP’’ (cVEMP) are a clinical demonstration of vestib- suggestive of vestibular dysfunction at neurological examination.
ulo-collic reflex (Rosengren et al., 2010; Brantberg, 2009; Park Symptoms and signs within 3 months before VEMP testing were
et al., 2010). The cVEMP pathway is believed to originate in the sac- considered as current, and those appearing more than 3 months
cular macula and continues through the vestibular nerve and previously at any time during the course of the disease, and that
nucleus, the vestibulospinal tracts, spinal motor nucleus and SCM improved and were not observed within the previous 3 months
muscles (Rosengren et al., 2010; Park et al., 2010; Bektas et al., were considered as past brainstem and cerebellar clinical
2008). cVEMP responses are characterized by biphasic waves with involvement.
initial positivity (p13) followed by a negative wave (n23). Recently, All previous and current MRI scans of each patient performed
myogenic response recorded from contralateral extraocular since the onset of the disease were reviewed for current and past
muscles in response to acoustic stimuli has been reported to be a brainstem or cerebellar lesions. Twenty patients who did not have
manifestation of crossed vestibulo-ocular reflex and named ‘‘ocular an MRI within 3 months before VEMP testing were excluded. Brain
VEMP’’ (oVEMP). The oVEMP pathway is thought to travel through MRI scans were evaluated by another examiner blinded to patients’
the medial longitudinal fasciculus (MLF), the oculomotor nuclei previous and current clinical findings. Brain lesions were described
and nerves and the extraocular muscles after the activation of as being localized to the right or left mesencephalon, pons, medulla
the vestibular nerve and nucleus (Rosengren et al., 2010). oVEMP and cerebellum. Brain lesions seen at MRI within 3 months before
responses are characterized by biphasic waves with an initial neg- VEMP testing were considered as current lesions. Brain lesions
ative peak (n1) followed by a positive peak (p1). seen on previous MRI scans but not observed on MRI scans within
The cVEMP test has become an important diagnostic tool, par- the previous 3 months were considered as past lesions.
ticularly in the evaluation of peripheral vestibular disorders.
VEMPs are thought to provide useful information about brainstem 2.3. VEMP recordings
functions, as the neural pathway of both VEMPs pass through the
brainstem. While cVEMP descends via the vestibulospinal tract VEMP tests were performed on the same day as the clinical
through the lower brainstem, oVEMP ascends via the MLF through examination by a different examiner blinded to the groups and pa-
the upper brainstem (Rosengren et al., 2007, 2010; Itoh et al., 2001; tients’ clinical and MRI examinations.
Tu and Young, 2004; Eleftheriadou et al., 2009; Lin et al., 2010). VEMP recordings were performed using a Medelec Synergy
Several studies have described cVEMP abnormalities in brain- EMG/EP machine (Oxford Instruments Medical, Surrey, UK). Pa-
stem lesions, including research with MS patients (Itoh et al., tients were tested in a sitting position. EMG signals were amplified
2001; Versino et al., 2002; Patko et al., 2007; Shimizu et al., and band pass-filtered between 1 and 1000 Hz. We used sound
2000; Sartucci and Logi, 2002; Pollak et al., 2006; Bandini et al., stimuli presented through headphones as rarefaction clicks of
2004; Alpini et al., 2004; Aidar and Suzuki, 2005). There are only 0.1 ms duration and a frequency of 5 Hz. A total of 128 stimuli
a few reports about the diagnostic value of oVEMP in brainstem were applied to each ear and repeated two consecutive times at
lesions (Rosengren et al., 2007; Rosengren and Colebatch, 2011; an intensity of 105 dB nHL. Click stimulus was preferred over
Su and Young, 2011). But none has evaluated the correlation of low frequency tone-burst due to the problems in generating short
oVEMP tests with both current and past clinical and MRI findings tone burst in our laboratory.
of MS patients. For the cVEMP test, active electrode was placed on the upper
We aimed to investigate the clinical value of ocular and cervical one-third of the SCM ipsilateral to the sound stimulation, with
VEMP tests in the evaluation of brainstem involvement in MS pa- the reference electrode on the anterior margin of the clavicle and
tients and their value in detecting silent brainstem lesions and to the ground electrode on the forehead. Patients were asked to turn
assess their relations with clinical and cranial MRI findings. their heads contralaterally to activate the ipsilateral SCM muscle
and to hold this position throughout the recording period. Muscle
activation was monitored during the recording and maintained at a
2. Methods
constant level (>50 lV). Peak latencies of p13 and n23 and peak-to-
peak amplitudes (p13-n23) were measured. The interside differ-
2.1. Subjects
ences of p13 and n23 latency and amplitude asymmetry ratio
(AR) were calculated. AR was calculated as follows: (larger
Sixty-two patients with definite MS according to the McDonald
response-smaller response)/(larger response + smaller response) 
criteria and 35 healthy volunteers were included (Polman et al.,
100 (Rosengren et al., 2010) We preferred to use AR for the inter-
2005). All patients were examined using otoscopy and audiometric
pretation of the VEMP amplitude, since VEMP response amplitude
testing before the study. Subjects with abnormal audiometric tests
is significantly affected by the force of muscular contraction or
and limited neck movements were excluded. The ethical commit-
stimulus intensity and exhibits wide variation.
tee approved the study, and informed consent was obtained from
For the oVEMP test, active electrode was placed around 1 cm
each subject.
below the center of the inferior eyelid contralateral to the sound
stimulation, with the reference electrode 15 mm below the ac-
2.2. Clinical and MRI examinations tive one and the ground electrode on the forehead. During the
test, patients were asked to look upward to a fixed point 2 m
All patients were questioned and examined for current brain- distant and 30–35° above the horizontal line. The peak latencies
stem or cerebellar clinical involvement. Previous brainstem or cer- of n1, p1 and peak-to-peak amplitudes (n1-p1) were measured.
ebellar clinical involvement of all patients was noted from their The interside differences of n1 and p1 latency and AR were
medical records. calculated.
Patients were then divided into groups according to the pres-
ence of current or past brainstem and cerebellar clinical involve- 2.4. Statistical analysis
ment. Both symptoms and signs were taken into consideration
while grouping patients. Patients with vertigo were grouped Statistical analysis was performed using SPSS for Windows ver-
according to the presence of additional brainstem or cerebellar sion 15. Means and standard deviations of each VEMP parameter
clinical findings. No additional test (caloric test, head shaking test, were determined. After evaluation of the assumption of the normal
etc.) was required since none of the patients had any findings distribution, student’s t test was applied to compare continuous
1874 S. Gazioglu, C. Boz / Clinical Neurophysiology 123 (2012) 1872–1879

variables and chi-square was applied for comparison of the cate- tude asymmetry ratios or interpeak latency differences of either
gorical variables. Correlations were tested using Pearson’s correla- ocular or cervical VEMPs. Figs. 1 and 2 illustrate examples of
tion analysis. Statistical significance was set at a p value of 0.05. oVEMP and cVEMP recordings in a control subject and a MS patient
with brainstem lesion. Amplitude versus latency scatter plots of
3. Results ocular and cervical VEMP responses of MS patients and control
group are illustrated in Fig. 3.
The MS group consisted of 40 females and 22 males (aged The upper limits of normal values (mean plus 2.5 SD) of each
16–55; mean age 34.7 ± 9.5) and the control group of 24 females VEMP parameter determined in the control group were 9.29 ms
and 11 males (aged 21–54; mean age 33.6 ± 9.7). There were no for n1 latency, 14.4 ms for p1 latency, 14.9 ms for p13 and
significant differences in age or gender among the groups 25.5 ms for n23 latency. The number of abnormal ocular or cervical
(p = 0.59 for age and p = 0.85 for gender). Fifty-five patients had VEMP responses compared by considering latency and amplitude
relapsing remitting MS (RRMS), 3 had primary progressive MS abnormalities on both sides in each group. Prolonged latencies or
(PPMS) and 4 had secondary progressive MS (SPMS). Median dura- absent responses were considered as abnormal. Only absence of
tion of the disease was 60 months (mean 75.8 ± 70.7) and mean any recording at all has been regarded as an amplitude abnormal-
expanded disability status scale (EDSS) score was 2.6 ± 2. Demo- ity. Interpeak latency differences, interaural latency differences
graphic features of the study population are shown in Table 1. and abnormalities in AR were not used as VEMP abnormality on
their own.
oVEMP abnormalities were seen in 45% (28/62) and cVEMP
3.1. VEMP abnormalities
abnormalities in 18% (11/62) of the MS patients. Combined assess-
ment of both ocular and cervical VEMP responses resulted in a 50%
Both ocular and cervical VEMP responses were detectable from
abnormality rate in MS patients. Abnormalities in both VEMP re-
both sides in all 35 control patients. cVEMPs were detectable from
sponses were statistically significant when compared to the con-
both sides in all patients. oVEMPs were recordable from both sides
trol subjects (p < 0.01) (Table 3).
in 56 patients, from one side in 3 patients and were absent bilater-
ally in 3 patients. Average values of each VEMP parameter for the
3.2. Correlation with brainstem clinical abnormalities
patient and control groups are summarized in Table 2. There were
no significant differences between male and female subjects in
Twenty of the 62 MS patients (32%) had current and 13 (21%)
either group. oVEMP mean n1 and p1 latencies and cVEMP mean
past brainstem clinical involvement. Overall, 53% (33/62) of pa-
p13 latency were significantly prolonged in MS patients (p <
tients had brainstem clinical involvement at least once during
0.01). But there were no significant differences in cVEMP mean
the course of their disease. The abnormality rate of combined
n23 latencies between MS patients and control subjects. Mean
VEMP responses in patients with current brainstem clinical
interaural latency differences for n1 and p1 latencies were signifi-
involvement was 50%. The abnormality rate of combined VEMP
cantly prolonged in MS patients (p < 0.01). But there was no signif-
responses in patients with past brainstem clinical involvement
icant difference in mean interaural latency difference for p13 and
was 61%. The abnormality rate of combined VEMP responses in pa-
n23 latencies. No significant differences were observed in ampli-
tients who had brainstem clinical involvement at least once during
the course of the disease was 54.5%. There was no statistically sig-
nificant correlation between the brainstem clinical and VEMP
Table 1 abnormalities. Among the 20 patients with current brainstem clin-
Demographic features of the study population. ical involvement, 4 had internuclear ophtalmoplegia (INO). The
Parameter MS patients Control subjects p value* oVEMP responses of 2 patients with INO were abnormal while
(n = 62) (n = 35) their cVEMP responses were normal. The other 2 patients with
Mean age (years) 34.7 ± 9.5 33.6 ± 9.7 0.59 INO had bilateral normal responses. According to the medical
Gender (female/male) 40/22 24/11 0.85 records, none of the patients presented with INO as a past brain-
Disease duration (month) 75.8 ± 70.7 stem clinical involvement.
EDSS 2.6 ± 2
*
p value; from t-test for age, from chi-square test for gender, significant if <0.05. 3.3. Correlation with brainstem MRI abnormalities

Eighteen out of the 62 MS patients (29%) had current and 15

(24%) had past brainstem lesions at MRI. Among the 18 patients
Table 2 with current brainstem MRI lesions, 12 had single brainstem le-
VEMP results of MS patients and controls. sions (1 in the mesencephalon, 1 in the medulla and 10 in the
Parameter MS patients Control subjects p value* pons) and 6 had lesions at more than one brainstem level (4 bilat-
eral, 2 unilateral). Among the 15 patients with past brainstem
oVEMP
Latency n1 (ms) 8.85 ± 1.23 8.07 ± 0.49 <0.01 MRI lesions, 8 had single brainstem lesions (all in the pons) and
Latency p1 (ms) 13.14 ± 1.55 12.20 ± 0.88 <0.01 7 had lesions at more than one brainstem level (5 bilateral, 2 uni-
n1-p1 interpeak latency 4.28 ± 1.01 4.13 ± 0.86 0.29 lateral). Details of the current brainstem lesions and the results of
Interaural latency diff. n1 0.6 ± 0.65 0.24 ± 0.32 <0.01
VEMP responses are shown in Table 4.
Interaural latency diff. p1 1.02 ± 0.91 0.44 ± 0.38 <0.01
Amplitude asymmetry ratio 23.41 ± 19.63 20.55 ± 15.74 0.46
The abnormality rate of combined VEMP responses in pa-
tients with current brainstem lesions was 56% (10/18). The
cVEMP
Latency p13 (ms) 12.80 ± 1.74 12.25 ± 1.06 <0.01
abnormality rate of combined VEMP responses in patients with
Latency n23 (ms) 21.34 ± 2.36 20.80 ± 1.88 0.1 past brainstem lesions was 67% (10/15). The abnormality rate
p13-n23 interpeak latency 8.5 ± 1.49 8.5 ± 1.49 0.98 of combined VEMP responses in patients who had brainstem
Interaural latency diff. p13 0.95 ± 0.73 1.11 ± 0.86 0.35 lesions at least once during the course of the disease was
Interaural latency diff n23 1.53 ± 1.15 1.22 ± 1.05 0.2
59.4%. There was no statistically significant correlation between
Amplitude asymmetry ratio 17.9 ± 14.1 15.6 ± 13.7 0.45
the VEMP abnormalities and brainstem lesions at MRI. Fig. 4
*
p value from t-test, significant if <0.05. illustrates the abnormality ratios of VEMP responses according
 S. Gazioglu, C. Boz / Clinical Neurophysiology 123 (2012) 1872–1879 1875

Fig. 1. Normal ocular and cervical VEMPs in a control patient.

Fig. 2. Ocular and cervical VEMP responses in a patient with right pontine MS plaque. Note the prolonged latencies (longer on the right) in oVEMP response and prolonged
latencies of right cVEMP response. T2-weighted MRI of the patient demonstrates a high signal lesion in the right pons.

to the presence of current brainstem clinical or MRI lesions. MS patients (13%) had current and 5 (8%) had past cerebellar le-
There was also no statistically significant correlation between sions at MRI. Most of the patients with cerebellar clinical
VEMP abnormalities being ipsilateral or contralateral to the involvement or MRI lesions also had brainstem clinical involve-
side of the brainstem MRI lesions. ment and MRI lesions at the same time. Only 2 patients showed
isolated cerebellar clinical involvement, and both their VEMP re-
3.4. Correlation with cerebellum sponses were normal. No patients had isolated cerebellar lesions
at MRI. Statistical correlations of cerebellar clinical involvement
Eleven out of the 62 MS patients (18%) had current and 4 and MRI lesions with VEMP responses were not therefore
(7%) past cerebellar clinical involvement. Eight out of the 62 performed.
1876 S. Gazioglu, C. Boz / Clinical Neurophysiology 123 (2012) 1872–1879

Fig. 3. oVEMP n1-p1 amplitude versus n1 latency and cVEMP p13-n23 amplitude versus p13 latency scatter plots of MS patients and the control group. Left and right sided
responses are combined. Vertical dashed lines indicate mean +2.5 SD. Note the prominent latency prolongations in MS patients compared to the control group in both ocular
and cVEMP responses.

Table 3
Abnormal VEMP responses in each group. these 9 patients with normal brainstem MRI exhibited abnormal
oVEMP cVEMP
combined VEMP responses.
Normal Abnormal Normal Abnormal
4. Discussion
MS patients (n = 62) 34 28 51 11
Control subjects (n = 35) 32 3 34 1
p value* p < 0.001 p < 0.05 The aim of this study was to investigate the clinical value of
* VEMP tests in the evaluation of the brainstem involvement in MS
p value from chi-square test, significant if <0.05.
patients. We found higher abnormality ratios of both VEMPs in
3.5. Correlation with duration of the disease and EDSS MS patients compared to the controls, but no correlation was ob-
served with the brainstem clinical or MRI findings.
cVEMP p13 and n23 latencies were significantly correlated Evoked potential latency prolongations suggestive of a demye-
with EDSS (r = 0.51, p < 0.01 for p13 and r = 0.4, p < 0.01 for n23. linating process are often seen in MS patients, and prolonged laten-
oVEMP n1 and p1 latencies also exhibited significant but weaker cies of VEMP responses in MS patients have been reported in many
correlation with EDSS (r = 0.28, p < 0.05 for n1 and r = 0.27, previous studies (Comi et al., 1999; Rosengren et al., 2007; Elefthe-
p < 0.05 for p1) Fig. 5 illustrates the correlation between ocular riadou et al., 2009; Shimizu et al., 2000; Sartucci and Logi, 2002;
and cervical VEMP latencies and EDSS. But there was no significant Bandini et al., 2004; Rosengren and Colebatch, 2011; Murofushi
correlation between any of the VEMP parameters and duration of et al., 2001). Our results also showed prolonged latencies of both
the disease. VEMP responses in MS patients compared to the controls. Only
the prolongation of n23 latency of cVEMP response was statisti-
cally insignificant. This is probably due to larger variation of n23
latency than p13 latency, which also has been observed in previous
3.6. Combined use of VEMP and MRI studies (Murofushi et al., 2001; Patko et al., 2007; Sartucci and
Logi, 2002). Prolonged p13 latency is reported to be the most com-
Patients with no brainstem clinical involvement at current eval- mon abnormality of cVEMP response in MS patients, similarly to
uation had 17% (7/42) abnormality rates in MRI, 43% (18/42) in our results. (Eleftheriadou et al., 2009; Patko et al., 2007; Shimizu
oVEMP, 14% (6/42) in cVEMP. Combined assessment of ocular et al., 2000; Sartucci and Logi, 2002; Murofushi et al., 2001). Delays
and cervical VEMPs raised the abnormality rate to 50% (21/42) of latency in evoked potentials are not specific for MS and do not
and combined assessment of VEMPs and MRI raised the abnormal- help to distinguish MS from other etiologies (Chiappa, 1984; Comi
ity rate to 57% (24/42). Both VEMPs were normal in 21 out of 42 et al., 1999; Eleftheriadou et al., 2009; Bandini et al., 2004; Alpini
patients with no brainstem clinical involvement. Three of these et al., 2004). VEMP latency prolongations have been reported in
21 patients exhibited brainstem lesion at MRI. other pathologies affecting the brainstem such as stroke and
Ten out of 20 patients with current brainstem clinical involve- tumors (Murofushi et al., 2001; Itoh et al., 2001; Su and Young,
ment exhibited normal combined VEMP responses. Five of these 2011). Interaural latency differences of VEMP responses may be
10 patients (50%) with normal VEMP responses exhibited brain- helpful in cases of unilateral demyelination when considered
stem lesions at MRI. There were no brainstem lesions at MRI in 9 together with the latency prolongations. The probable reason for
of the 20 patients with brainstem clinical involvement. Four of no significant difference being observed in mean interaural latency
 S. Gazioglu, C. Boz / Clinical Neurophysiology 123 (2012) 1872–1879 1877

Table 4 were higher than the values reported in other studies. This may be
Details of the current brainstem lesions and VEMP results. due to our laboratory settings, degree of electrode separation, mus-
Patient Brainstem lesion oVEMP cVEMP cle contraction or stimulus intensity. Patko et al. reported signifi-
no. cantly lower amplitudes of cVEMP responses in MS patients
1 Bilateral mesencephalon, Bilateral Abnormal on (Patko et al., 2007). But they did not compare AR. Bandini et al.
pons, medulla abnormal the right reported that cVEMP amplitude values make no significant diag-
2 Bilateral mesencephalon, Bilateral Bilateral nostic contribution in MS patients (Bandini et al., 2004).
pons abnormal abnormal
3 Bilateral pons, medulla Bilateral normal Bilateral normal
We observed a 17.7% abnormality rate in cVEMP responses in
4 Bilateral pons, medulla Bilateral normal Bilateral normal MS patients. Previous studies have reported these at levels ranging
5 Right mesencephalon, pons, Bilateral normal Bilateral normal from 31% to 60% (Eleftheriadou et al., 2009; Versino et al., 2002;
medulla Patko et al., 2007; Sartucci and Logi, 2002; Bandini et al., 2004).
6 Right mesencephalon Bilateral normal Bilateral normal
The lower abnormality level of cVEMP responses in our study
7 Right pons Bilateral normal Bilateral normal
8 Right pons Bilateral normal Bilateral normal may be due to different study populations and upper limit values.
9 Right pons Bilateral Bilateral normal In addition, using only the absence of the responses as an ampli-
abnormal tude abnormality and not defining a limit value for the interpreta-
10 Right pons Abnormal on Bilateral normal tion of the amplitudes may have contributed to this low
the right
abnormality level of cVEMP responses in our study. The abnormal-
11 Right pons Abnormal on Abnormal on
the right the right ity ratio of oVEMP responses (45.2%) was higher than that of the
12 Right medulla Bilateral Bilateral normal cVEMP responses in our study. Our results support the hypothesis
abnormal that oVEMP is more sensitive than cVEMP due to the higher inci-
13 Left pons, medulla Abnormal on Bilateral
dence of medial longitudinal fasciculus (MLF) lesions in MS.
the left abnormal
14 Left pons Bilateral normal Bilateral normal (Rosengren et al., 2007). Using ocular and cervical VEMPs together
15 Left pons Abnormal on Bilateral normal permits the evaluation of both ascending and descending vestibu-
the right lar pathways in the brainstem, thus resulting in a high percentage
16 Left pons Bilateral normal Bilateral normal of abnormality (50%) (Rosengren et al., 2007, 2010; Tu and Young,
17 Left pons Bilateral Bilateral normal
2004; Lin et al., 2010). Among the other evoked potentials, visual
abnormal
18 Left pons Bilateral Abnormal on evoked potential (VEP) abnormalities have been reported to be
abnormal the left higher in MS patients due to the greater frequency of optic neurop-
athy (Chiappa, 1984). The rate of VEMP abnormalities observed in
our study is close to the abnormality levels of brainstem auditory
for p13 and n23 latencies, although there was a statistically signif- evoked potential (BAEP) evaluating brainstem pathways such as
icant difference in mean interaural latency for n1 and p1 between VEMP, which have been reported at between 30% and 64% in pre-
the MS patients and the controls in our study, is that cVEMP abnor- vious studies (Chiappa, 1984; Comi et al., 1999; Patko et al., 2007;
malities are at a lower level compared to oVEMP abnormalities. Iriarte et al., 1991; David et al., 1990; Ko, 2010).
Sartucci et al. reported that interside asymmetry did not make as Patients who had brainstem clinical involvement or MRI lesions
much of a clinical contribution in VEMP as in other evoked poten- at any time during the course of the disease showed higher abnor-
tials (Sartucci and Logi, 2002). mality rates than patients with no brainstem clinical involvement
There were no significant differences in amplitude AR of either or MRI lesions in either ocular or cervical VEMP responses, oVEMP
ocular or cervical VEMPs. Although this finding may suggest that abnormalities being higher than cVEMP abnormalities. But the dif-
the amplitude values of VEMP responses are of no very significant ference was not statistically significant among the groups. Several
diagnostic value in MS patients, amplitudes of VEMP responses studies have discussed the relation between VEMP abnormalities
should not be used alone as an abnormality criterion and should and brainstem involvement. Rosengren et al. determined a high le-
be interpreted together with latency values in MS patients due to vel of oVEMP abnormality in 12 patients with INO, the great major-
being so much influenced by factors such as muscle contraction ity of whom consisted of MS subjects. However, they performed no
or stimulus intensity. Amplitudes of oVEMP responses in our study comparison with a patient population with no brainstem clinical

Fig. 4. Abnormality ratios of VEMP responses according to the presence of current brainstem clinical or MRI lesions.
1878 S. Gazioglu, C. Boz / Clinical Neurophysiology 123 (2012) 1872–1879

Fig. 5. The correlation between between ocular and cervical VEMP latencies and EDSS. Left and right sided responses are combined. Both ocular and cervical VEMP latencies
exhibited a positive correlation with EDSS (r = 0.51, p < 0.01 for p13 and r = 0.28, p < 0.05 for n1).

findings (Rosengren and Colebatch, 2011). We also determined between VEMP abnormality and the duration of the disease in con-
abnormal ocular but normal cervical VEMP responses in 2 of our trast to the literature. Significant correlation between duration of
4 patients with INO indicating the value of oVEMP responses in the disease and abnormalities of cVEMP responses have been
the assessment of patients with lesions of internuclear pathways. reported in previous studies (Patko et al., 2007; Bandini et al.,
Although no statistical correlation could be established due to 2004). The high percentage of patients with short disease duration
the small number of cases with INO in our study, future studies in our study population may be responsible for this result.
with a greater number of MS patients with and without INO may Forty-six percent of our patients had disease duration of less than
be useful in corroborating these data. In a study of 70 MS patients, 5 years, and only 19% of greater than 10 years.
Versino et al. reported that brainstem clinical findings were not In conclusion, although it is not specific, VEMP testing, and
correlated with cVEMP abnormalities (Versino et al., 2002). Patko especially the oVEMP test, provides useful information about
et al. reported in a study of 30 MS patients that cVEMP abnormal- brainstem involvement in MS patients. The combination of VEMPs
ities were not correlated with clinical findings but were highly with other evoked potentials permits more comprehensive evalu-
correlated with MRI (Patko et al., 2007). Other previous studies ation of the brainstem pathways in MS patients. We think that
with cVEMP also reported no significant correlation between the value of VEMPs in detecting silent brainstem lesions as in other
cVEMP abnormalities and brainstem MR lesions, similarly to our evoked potentials, the opportunity to evaluate both upper and
results. (Bandini et al., 2004; Pollak et al., 2006; Eleftheriadou lower brainstem using the combination of both VEMPs and the cor-
et al., 2009). Abnormal MRI signals in the brainstem in a MS patient relation with disability support the clinical usefulness of the VEMP
sometimes may be due to inflammation and edema rather than test as a new additional neurophysiological test in the evaluation
demyelination, and may cause subtle lesions to appear larger in of MS patients.
MRI. Partly affected fibers due to these lesions are sometimes
insufficient to produce the expected abnormality in conduction References
(Comi et al., 1999; Patko et al., 2007; Bandini et al., 2004; Ross
et al., 1992). In contrast, small demyelinating lesions in the brain- Aidar RC, Suzuki FA. Vestibular evoked myogenic potential: new perspectives in
stem that cannot be detected by MRI can produce conduction multiple sclerosis. Braz J Otorhinolaryngol 2005;71:48–54.
Akin FW, Murnane OD, Panus PC, Caruthers SK, Wilkinson AE, Proffitt TM. The
abnormality, and this reflects the value of evoked potentials in influence of voluntary tonic EMG level on the vestibular-evoked myogenic
detecting silent lesions (Comi et al., 1999; Patko et al., 2007; potential. J Rehabil Res Dev 2004;41(3B):473–80.
Bandini et al., 2004; Rosengren and Colebatch, 2011; Gronseth Alpini D, Pugnetti L, Caputo D, Cornelio F, Capobianco S, Cesarani A. Vestibular
evoked myogenic potentials in multiple sclerosis: clinical and imaging
and Ashman, 2000). Abnormalities of evoked responses in asymp- correlations. Mult Scler 2004;10:316–21.
tomatic MS patients have been reported previously (Comi et al., Bandini F, Beronio A, Ghiglione E, Solaro C, Parodi RC, Mazzella L. The diagnostic
1999; Fuhr and Kappos, 2001; Emerson, 1998). Santos et al. ob- value of vestibular evoked myogenic potentials in multiple sclerosis. J Neurol
2004;251:617–21.
served 58.62% abnormalities in BAEP in patients with MS without Bektas D, Gazioglu S, Arslan S, Cobanoglu B, Boz C, Caylan R. VEMP responses are not
brainstem involvement (Santos et al., 2003). Our results indicate affected in non-insulin-dependent diabetes mellitus patients with or without
that VEMP, a test more easily performed than BAEP, is also able polyneuropathy. Acta Otolaryngol 2008;128:768–71.
Brantberg K. Vestibular evoked myogenic potentials (VEMPs): usefulness in clinical
to show silent brainstem lesions in MS patients, oVEMP being more
neurotology. Semin Neurol 2009;29:541–7.
sensitive than cVEMP. Cherchi M, Bellinaso NP, Card K, Covington A, Krumpe A, Pfeifer MS, et al. Sound
We determined a significant correlation between VEMP abnor- evoked triceps myogenic potentials. Otol Neurotol 2009;30:545–50.
malities and the level of the disability caused by the disease Chiappa KH. Pattern-shift visual, brainstem auditory and short-latency
somatosensory evoked potentials in multiple sclerosis. Ann N Y Acad Sci
detected by EDSS. Bandini et al. also reported a significant correla- 1984;436:315–27.
tion between cVEMP latency changes and disability measures Comi G, Leocani L, Medaglini S, Locatelli T, Martinelli V, Santuccio G, et al.
(Bandini et al., 2004). This finding suggests that progressive demy- Measuring evoked responses in multiple sclerosis. Mult Scler 1999;5:263–7.
David P, Ristori GP, Elia M, Bartoli A, Ciervo A, Massaro AR et al. Multiple sclerosis.
elination with advancing disease leads more often to abnormalities Magnetic resonance imaging, evoked potentials and cerebrospinal fluid
in VEMP responses. We established no significant correlation analysis. Acta Neurol (Napoli) 1990;12:200–6.
 S. Gazioglu, C. Boz / Clinical Neurophysiology 123 (2012) 1872–1879 1879

Debatisse D, Pralong E, Guerit JM, Bisdorff A. Recording click-evoked myogenic Pollak L, Kushnir M, Stryjer R. Diagnostic value of vestibular evoked myogenic
potentials (CEMPs) with a setup for brainstem auditory evoked potentials potentials in cerebellar and lower-brainstem strokes. Neurophysiol Clin
(BAEPs). Neurophysiol Clin 2005;35:109–17. 2006;36:227–33.
Eleftheriadou A, Deftereos SN, Zarikas V, Panagopoulos G, Sfetsos S, Karageorgiou Polman CH, Reingold SC, Edan G, Filippi M, Hartung HP, Kappos L, et al. Diagnostic
CL, et al. The diagnostic value of earlier and later components of Vestibular criteria for multiple sclerosis: 2005 revisions to the ‘‘McDonald Criteria’’. Ann
Evoked Myogenic Potentials (VEMP) in multiple sclerosis. J Vestib Res Neurol 2005;58:840–6.
2009;19:59–66. Rosengren SM, Nogajski JH, Cremer PD, Colebatch JG. Delayed vestibular evoked
Emerson RG. Evoked potentials in clinical trials for multiple sclerosis. J Clin responses to the eyes and neck in a patient with an isolated brainstem lesion.
Neurophysiol 1998;15:109–16. Clin Neurophysiol 2007;118:2112–6.
Fuhr P, Kappos L. Evoked potentials for evaluation of multiple sclerosis. Clin Rosengren SM, Welgampola MS, Colebatch JG. Vestibular evoked myogenic
Neurophysiol 2001;112:2185–9. potentials: past, present and future. Clin Neurophysiol 2010;121:636–51.
Gronseth GS, Ashman EJ. Practice parameter: the usefulness of evoked potentials in Rosengren SM, Colebatch JG. Ocular vestibular evoked myogenic potentials are
identifying clinically silent lesions in patients with suspected multiple sclerosis abnormal in internuclear ophthalmoplegia. Clin Neurophysiol 2011;122:
(an evidence-based review): report of the Quality Standards Subcommittee of 1264–7.
the American Academy of Neurology. Neurology 2000;54:1720–5. Ross MA, Leis AA, Krain L, Mitchell G. Normal conduction in pathways traversing an
Iriarte J, de Castro P, Gudin M, Martinez-Lage JM. Comparative study of magnetic asymptomatic multiple sclerosis plaque. Electroencephalogr Clin Neurophysiol
resonance and evoked potentials in patients with clinically defined multiple 1992;85:42–5.
sclerosis. Neurologia 1991;6:322–7. Santos MA, Peixoto MA, Munhoz MS, de Almeida AV. Auditory evoked potentials
Itoh A, Kim YS, Yoshioka K, Kanaya M, Enomoto H, Hiraiwa F, et al. Clinical study of evaluation of the brain stem in multiple sclerosis. Arq Neuropsiquiatr
vestibular-evoked myogenic potentials and auditory brainstem responses in 2003;61:392–7.
patients with brainstem lesions. Acta Otolaryngol Suppl 2001;545:116–9. Sartucci F, Logi F. Vestibular-evoked myogenic potentials: a method to assess
Ko KF. The role of evoked potential and MR imaging in assessing multiple sclerosis: vestibulo-spinal conduction in multiple sclerosis patients. Brain Res Bull
a comparative study. Singapore Med J 2010;51:716–20. 2002;59:59–63.
Lin KY, Hsu YS, Young YH. Brainstem lesion in benign paroxysmal vertigo children: Shimizu K, Murofushi T, Sakurai M, Halmagyi M. Vestibular evoked myogenic
evaluated by a combined ocular and cervical vestibular-evoked myogenic potentials in multiple sclerosis. J Neurol Neurosurg Psychiatry 2000;69:276–7.
potential test. Int J Pediatr Otorhinolaryngol 2010;74:523–7. Su CH, Young YH. Differentiating cerebellar and brainstem lesions with ocular
Murofushi T, Shimizu K, Takegoshi H, Cheng PW. Diagnostic value of prolonged vestibular-evoked myogenic potential test. Eur Arch Otorhinolaryngol
latencies in the vestibular evoked myogenic potential. Arch Otolaryngol Head 2011;268:923–30.
Neck Surg 2001;127:1069–72. Tu CE, Young YH. Audiovestibular evolution in a patient with multiple sclerosis. Ann
Park HJ, Lee IS, Shin JE, Lee YJ, Park MS. Frequency-tuning characteristics of cervical Otol Rhinol Laryngol 2004;113:726–9.
and ocular vestibular evoked myogenic potentials induced by air-conducted Versino M, Colnaghi S, Callieco R, Bergamaschi R, Romani A, Cosi V. Vestibular
tone bursts. Clin Neurophysiol 2010;121:85–9. evoked myogenic potentials in multiple sclerosis patients. Clin Neurophysiol
Patko T, Simo M, Aranyi Z. Vestibular click-evoked myogenic potentials: sensitivity 2002;113:1464–9.
and factors determining abnormality in patients with multiple sclerosis. Mult Zhou G, Cox LC. Vestibular evoked myogenic potentials: history and overview. Am J
Scler 2007;13:193–8. Audiol 2004;13:135–43.