Vestibular evoked myogenic potentials (VEMP) are myogenic

ORIGINAL RESEARCH Ocular and Cervical Vestibular Evoked Myogenic Potentials in Patients With Multiple Sclerosis Tereza Gabelic,* Magdalena Krbot, Ana B. Sefer, Velimir Isgum,* Ivan Adamec,* and Mario Habek* Objectives: The aim of this study was to evaluate latencies and corrected p13-n23 cervical vestibular evoked myogenic potentials (cvemp) and n10-p13 ocular vestibular evoked myogenic potentials (ovemp) amplitudes in patients with relapsing remitting multiple sclerosis (MS). Methods: This was a prospective, case-control study. Thirty patients with MS and 15 healthy controls were included. Cervical vestibular evoked myogenic potentials and ovemp in response to acoustic clicks of 1 ms duration at the intensity of 130 db SPL and the stimulation frequency of 1 Hz were studied. Signals were divided in segments of 120 ms duration (20 ms before the stimulus and 100 ms after the stimulus) and averaged. Results: In MS group, there was significant latencies prolongation of all sternocleidomastoid responses (p13 and n23) and n10 response of the ocular muscles. The sternocleidomastoid p13-n23 normalized amplitude was significantly higher in MS patients. Prolonged latencies were found in 57% and conduction block in 7% of patients in at least one sternocleidomastoid response in the MS group. Prolonged latencies were found in 30% and conduction block in 40% of patients in at least one ocular response in the MS group. When cvemp and ovemp are combined, 80% had pathological finding. When correlating brainstem clinical, brainstem MRI, and cvemp findings, there was no statistical significance (brainstem clinical vs. cvemp P ¼ 0.1; brainstem MRI vs. cvemp P ¼ 0.82). When correlating brainstem clinical, brainstem MRI and ovemp findings, there was a statistical significant correlation between brainstem clinical versus ovemp, P ¼ 0.02, whereas there was no statistical significance between brainstem MRI versus ovemp (P ¼ 0.38). Conclusions: Combination of cvemp and ovemp in MS patients allows better estimation of brainstem lesions. Key Words: Cervical vestibular evoked myogenic potentials, Ocular vestibular evoked myogenic potentials, Amplitudes, Latencies, Multiple sclerosis. (J Clin Neurophysiol 2013;30: 86 91) Vestibular evoked myogenic potentials (VEMP) are myogenic short latency responses evoked by sound- or bone-conducted impulse. This diagnostic method is in use less than 30 years and together with posturography is fundamental test assessing vestibulospinal involvement (Rosengren et al., 2010). Neuroanatomical basis of this test is based on stimulation of vestibular system by acoustic, From the *Referral Center for Demyelinating Diseases of the Central Nervous System, Department of Neurology, University Hospital Center Zagreb, Zagreb, Croatia; University of Zagreb, Zagreb, Croatia; and Department of Neurology, School of Medicine, Univerisity of Zagreb, Zagreb, Croatia. Study concept and design: Habek; Acquisition of data: Gabelic, Krbot, Sefer, Isgum, Adamec, and Habek; Analysis and interpretation of data: Gabelic, Krbot, Sefer, Isgum, Adamec, and Habek; Drafting of the manuscript: Gabelic; Critical revision of the manuscript for important intellectual content: Gabelic, Krbot, Sefer, Isgum, Adamec, and Habek; Administrative, technical, and material support: Gabelic, Krbot, Sefer, Isgum, Adamec, and Habek. Address correspondence and reprint requests to Mario Habek, MD, PhD, Department of Neurology, Zagreb School of Medicine and University Hospital Center, Kispaticeva 12, HR 10000 Zagreb, Croatia; e-mail: mhabek@mef.hr. Copyright Ó 2013 by the American Clinical Neurophysiology Society ISSN: 0736-0258/13/3001-0086 electric, or vibratory impulse. When stimulation is delivered, receptors in the sacculus and utriculous are activated. Impulse is further transmitted trough vestibular nerve and vestibular nuclei. Cervical VEMPs (cvemp) are manifestation of vestibulocolic reflex, and the impulse is transmitted through vestibulospinal pathways to ipsilateral sternocleidomastoid (SCM) muscle on which myogenic potentials are recorded. Therefore, cvemp evaluates integrity of vestibulospinal pathway. Ocular VEMPs (ovemp) are a manifestation of vestibuloocular reflex, and after impulse is transmitted to vestibular nuclei, it crosses contralaterally through medial longitudinal fasciculi to the extraocular muscle (Rosengren et al., 2010). These evoked potentials are of vestibular origin. Intense air-conducted sound elicits an ovemp, and it has been suggested that it does so by stimulating saccular receptors and afferents in the inferior vestibular nerve and so activating a crossed sacculoocular pathway. (Curthoys et al., 2011; Jacobson et al., 2011). These findings support the hypothesis that the ovemp in response to air-conducted sound predominantly reflects utricular functions while air-conducted sound cvemp reflects saccular functions (Jacobson et al., 2011; Murofushi et al., 2011). Vestibular evoked myogenic potential is a useful diagnostic method in an evaluation of clinically silent lesions in patients with multiple sclerosis (MS), especially in lesions of lower pons and medulla oblongata (Alpini et al., 2005; Versino et al., 2002). Abnormal results of VEMP in patients with MS implicates lesion of the brainstem, despite normal MRI and/or neurological examination (Eleftheriadou et al., 2009). Characteristic findings in MS patients are prolonged latencies, similar to as in visual evoked potentials or a nerve conduction block. Several studies investigated cvemp abnormalities in patients with MS, whereas ovemp abnormalities were investigated in only one study in patient with MS who had internuclear ophthalmoplegia, showing mostly conduction block or delayed latencies (Rosengren and Colebatch, 2011). The aim of this study was to evaluate latencies and normalized p13-n23 cvemp and n10-p13 ovemp amplitudes in patients with relapsing remitting MS. MATERIALS AND METHODS This was a prospective, case-control study. Patients with relapsing remitting MS, fulfilling McDonald criteria, were included (Polman et al., 2011). Normal controls were collected prospectively. Study was approved by the ethical committee of the University Hospital Center Zagreb. All participants (patients and normal controls) signed informed consent form after the experiment was explained to them in details. Methods of recordings and analysis of recorded data were designed according to previously described details (Rosengren et al., 2010). During the experiment, participants sat in comfortable chair. Patients were instructed to slightly move their head away from the back of the chair and push it forward to activate SCM muscle. The 86 Journal of Clinical Neurophysiology Volume 30, Number 1, February 2013

Journal of Clinical Neurophysiology Volume 30, Number 1, February 2013 VEMP in MS Patients contraction of muscle was maintained because of the cooperation of patients in maintaining the same position during the test. Participants were also instructed to direct their gaze to the ceiling to activate ocular muscles (OMs). The evoked response from the SCM was recorded from the active surface electrode placed on the belly of the SCM of the stimulated side and referred to the surface electrode placed on the tendon of the same SCM. The evoked response from the OM was recorded from two surface electrodes situated 2 cm below the contralateral eye. Active electrode was situated closer to the eye and referred to the reference 1 cm below. The stimuli were delivered by a pair of headphones in series of 50 trails to 1 ear at a time and repeated two times for each ear to provide reproducibility. The presented stimuli were acoustic clicks of 1 ms duration at the intensity of 130 db SPL and the stimulation frequency of 1 Hz. Recording were performed using a Brain Products Brain Vision Recorded, and the analysis of the recorded data was performed using a Brain Products Brain Vision Analyzer. Signals were filtered with bandpass filter from 5 to 1000 Hz. For the purpose of the analysis, signals were divided in segments of 120 ms duration (20 ms before the stimulus and 100 ms after the stimulus) and averaged for each set of 50 trials. From the averaged responses from the two sets, the grand average was computed and used for further analysis (as shown in Fig. 2 for one patient). We used baseline-normalized values of the SCM amplitude data instead of the absolute value of amplitude, because absolute amplitude of the evoked response depends on the amplitude of the muscle activity (muscle contraction) (Alpini et al., 2004) and is not reliable measure. The baseline-normalized value of amplitude is calculated by dividing the absolute peak to peak amplitude (p13-n23) with mean value of rectified activity of muscle in the period prior the stimulus. Statistic analysis was performed using IBM SPSS 19.0 (Chicago, IL). The difference between categorical variables was calculated by using x 2 test, and for the numeric variables, two-tailed Student t-test for independent samples was used. P values less than 0.05 were considered significant. RESULTS Thirty patients with MS and 15 healthy controls were included in the study. There were no significant differences between men and women and age in control and evaluated MS group. Clinical and MRI characteristics of the MS cohort are presented in Table 1. Only 12 of 30 (40%) MS patients had clinical signs of brainstem involvement, whereas 11 of 25 (44%) MS patients had brainstem pathology evident on the brain MRI. Most patients were treatment naive, and 12 (40%) were on immunomodulatory therapy (interferon beta/copaxone). TABLE 1. Sex/Age Clinical and MRI Characteristics of the MS Cohort EDSS Brainstem EDSS Brainstem Lesion Evident on Brain MRI Disease Duration (Years) cvemp ovemp F/44 2,5 0 2 1 1 1 M/44 2 2 1 7 1 2 M/31 2,5 2 2 2 1 2 F/29 1 1 2 1 1 1 M/31 2,5 2 1 0,2 1 2 M/47 0 0 2 0,2 2 1 M/22 3 0 1 3 1 1 M/21 3,5 0 UNK 8 1 1 F/36 2,5 2 2 1 2 2 M/26 2 2 UNK 3 2 2 F/31 0 0 2 1 1 1 F/23 3,5 2 1 0,2 1 1 M/36 1,5 0 2 0,1 1 1 M/20 0 0 1 0,2 2 2 F/39 2,5 0 1 2 2 2 M/26 3 2 1 2 1 2 F/48 2,5 0 1 13 2 1 F/48 5 0 UNK 20 1 1 M/42 1,5 0 2 1 2 1 M/26 4 3 UNK 6 1 1 F/28 2 0 2 0,1 2 2 F/26 2 0 2 4 2 1 M/24 2,5 2 1 1 2 2 M/39 2 0 UNK 3 1 1 M/50 3 0 2 2 2 1 F/22 3,5 0 2 3 2 1 F/21 2,5 0 2 1 1 1 F/29 2,5 2 1 3 1 1 F/48 4 0 1 21 2 1 M/44 3 2 2 8 1 2 M, male; F, female; EDSS, Expanded Disability Status Scale; UNK, MRI unavailable at the time of the VEMP; 1, pathological finding on MRI/VEMP; 2, normal finding on MRI/VEMP. Copyright Ó 2013 by the American Clinical Neurophysiology Society 87

T. Gabelic et al. Journal of Clinical Neurophysiology Volume 30, Number 1, February 2013 TABLE 2. Results of cvemp and ovemp Between Groups MS (Mean) Healthy Controls (Mean) SCM lat p13 15.8 ms 14.6 ms 0.002 SCM lat n23 24.9 ms 22.1 ms,0.001 SCM ampl p13-n23 1.9 mv 1.6 mv 0.018 O lat n10 10.3 ms 9.5 ms 0.009 O lat p13 14.9 ms 14.0 ms 0.052 O ampl n10-p13 13.1 mv 13.3 mv 0.947 SCM, strnocleidomastoid; O, ocular; lat, latency; ampl, amplitude. Regarding results of cvemp, altogether 17 patients (57%) had prolonged latencies, and conduction block was identified in 2 patients (7%) of at least 1 sternocleidomastoid response in the MS group. In addition, in the MS group, there was significant p13 and n23 latencies prolongation compared with controls. The p13-n23 normalized amplitude was significantly higher in MS patients compared with controls. When correlating brainstem clinical, P brainstem MRI, and cvemp findings, there was no statistical significance (brainstem clinical vs. cvemp, P ¼ 0.1; brainstem MRI vs. cvemp, P ¼ 0.82). Regarding results of ovemp, altogether 9 patients (30%) had prolonged latencies, and conduction block was identified in 12 patients (40%) of at least 1 ocular response in the MS group. In addition, latency of the n10 response showed significant prolongation in the MS group compared with controls. Statistical significance was not reached for p13 latencies of the OMs. There was no statistically significant difference for ocular n10-p13 amplitude between groups. When correlating brainstem clinical, brainstem MRI, and ovemp findings, there was a statistical significant correlation between brainstem clinical versus ovemp, P ¼ 0.02, whereas there was no statistical significance between brainstem MRI versus ovemp (P ¼ 0.38). When cvemp and ovemp are combined, 24 patients (80%) had pathological finding. Results of statistical analysis are shown in Table 2 in numerical form and in Fig. 1 in the form of a boxplot. In Fig. 2, tracings of cvemp and ovemp and, in Fig. 3, superimposed tracings form two repeated tests from one of the MS patients are shown. FIG. 1. Results of statistical analysis in form of a boxplot. Red, healthy control group; blue, MS group; SCM, sternocleidomastoid; O, ocular; lat, latency; ampl, amplitude. 88 Copyright Ó 2013 by the American Clinical Neurophysiology Society

Journal of Clinical Neurophysiology Volume 30, Number 1, February 2013 VEMP in MS Patients FIG. 2. Tracings of ocular and cervical vestibular evoked myogenic potentials in one of the studied patients showing delayed p13 and n23 latencies on right sternocleidomastoid muscles and high ocular n10-p13 amplitudes for both left and right ocular muscles. SCMR, right sternocleidomastoid; SCML, left sternocleidomastoid; OR, right ocular; OL, left ocular. FIG. 3. Superimposed tracings from two repeats of the ocular and cervical vestibular evoked myogenic potentials in the same patient, as in Fig. 2. SCMR, right sternocleidomastoid; SCML, left sternocleidomastoid; OR, right ocular; OL, left ocular. Copyright Ó 2013 by the American Clinical Neurophysiology Society 89

T. Gabelic et al. Journal of Clinical Neurophysiology Volume 30, Number 1, February 2013 DISCUSSION Results of our study showed significant prolongation of latencies for SCM muscles and for n10 response of OMs in patients with MS compared with healthy controls. In addition, ovemp in greater extent picks the brainstem pathology in MS than the clinical examination in a form of expanded disability status scale. In patients with MS, cvemps are abnormal in up to 50% of patients (Bandini et al., 2004; Murofushi et al., 2001; Patkó et al., 2007; Sartucci and Logi, 2002; Shimizu et al., 2000; Versino et al., 2002). Absence of response is usually caused by severe damage of the myelin sheaths, sometimes accompanied with axonal damage (Patkó et al., 2007; Versino et al., 2002). Latencies prolongations in cvemp are well known and established in many articles (Alpini et al., 2005; Bandini et al., 2004; Eleftheriadou et al., 2009; Patkó et al., 2007; Versino et al., 2002), but this is not the case with amplitude values. Amplitude decrease of cvemp was shown in few articles (Patkó et al., 2007; Sartucci and Logi, 2002; Versino et al., 2001; Versino et al., 2002) reaching statistical difference for mean raw p13-n23 amplitude but not in normalized amplitude values. One study showed increased p13-n 23 cvemp amplitude in MS patients (Aidar and Suzuki, 2005). In current study, latencies were found to be significantly prolonged for all SCM responses. In addition, we have shown significantly higher p13-n23 baseline normalized amplitude for the SCM muscles in MS patients compared with healthy controls. This is an interesting observation, pathophysiological mechanism of which is not fully understood. Some authors have suggested that patients with Tullio phenomenon have high cvemp amplitudes; however, the finding of a large VEMP amplitude in response to a high-intensity sound stimulation is not, per se, distinctive for a significant vestibular hypersensitivity to sounds (Brantberg et al., 2007). Conversely, finding of high amplitude of steady-state visual evoked potentials in patients with migraine suggests abnormal excitability in the primary visual cortex caused by an impairment of the inhibitory interneurons. (Shibata et al., 2008) The interpretation of VEMP aplitudes has another problem. In our study, we used baseline-normalized values of amplitudes and not absolute values of amplitudes. In studies demonstrating decreased amplitudes of VEMPs (Patkó et al., 2007; Versino et al., 2001; Versino et al., 2002), peak to peak amplitudes were measured and normalized value was not calculated, making them hard to interpret. It is necessary to standardize amplitude measurement and its calculation to avoid mistakes and differences that are possible when uniform values are not used. Another important aspect of evoked potentials is its ability to detect lesions that are not present clinically or on MRI. Concordance of cvemp with clinical findings of presence/absence of brainstem involvement was found in 55% and with MRI findings in 65% of the cases in one study (Alpini et al., 2004). However, in some patients, cvemps are normal despite a neuroradiologic evidence of demyelinating brainstem lesions (Alpini et al., 2004). The role of ovemp in MS has not been systematically investigated. We have found conduction block in 40% and prolonged latencies in 30% of the OM responses in MS patients and significantly prolonged latencies of the n10 response in MS group compared with controls. Amplitude analysis of OMs did not show statistically significant difference between groups. There is only 1 study investigating ovemp in 12 patients with internuclear ophthalmoplegia (in most patients because of MS). Nearly all patients (85%) had an ovemp abnormality on at least one side, while only 15% of patients had a cvemp abnormality (Rosengren and Colebatch, 2011). In the patients with a kinetic internuclear ophthalmoplegia, ovemps were present, although sometimes delayed, suggesting that more subtle clinical abnormalities are associated with only mild ovemp abnormalities. However, it is likely that the ovemp can also detect clinically silent brainstem lesions similar to cvemp (Bandini et al., 2004). We have recently reported a patient with clinically isolated syndrome suggestive of MS who presented with primary position upbeat nystagmus. MRI revealed a demyelinating lesion in the lower medulla, which affected the nucleus intercalatus; this type of lesion inhibits the flocculovestibular inhibitory pathway, thereby causing upbeat nystagmus. Nystagmus still persisted after pulse corticosteroid therapy. This could be due to a loss of central adaptation of the vestibuloocular system in our patient because of more diffuse brainstem damage shown on VEMPs, which showed delayed p13 and n23 latencies on both SCM muscles, prolonged n10 and p13 latencies on the right extraocular muscles, and a conduction block for the left extraocular muscles (Adamec et al., 2012). This case nicely shows how ovemp and cvemp can supplement the diagnostic procedures in MS patients and that ovemps may be valuable for detecting and confirming lesions of otolith ocular pathways. In conclusion, although MRI is considered to be the single most sensitive test for MS, it recognized that it cannot detect all lesions in MS patients. Thus, the development of a new, potentially diagnostic test could enhance the detection of functional central nervous system dysfunction. On the basis of our results, we find VEMP to be a quick, simple, painless, and reliable neurophysiological test in assessing lesions of the brainstem in patients with MS. For a definite evaluation of the role of the cvemps and ovemps, additional studies with larger number of patients are necessary. ACKNOWLEDGMENTS The authors thank Dr. Iva Hojsak for statistical analysis and constructive comments regarding the manuscript. REFERENCES Adamec I, Gabelic T, Krbot M, et al. Primary position upbeat nystagmus. J Clin Neurosci 2012;19:161 162. Aidar RC, Suzuki FA. Vestibular evoked myogenic potential: new perspectives in multiple sclerosis. Braz J Otorhinolaryngol 2005;71:48 54. Alpini D, Pugnetti L, Caputo D, Cesarani A. Vestibular evoked myogenic potentials in multiple sclerosis: a comparison between onset and definite cases. Int Tinnitus J 2005;11:48 51. Alpini D, Pugnetti L, Caputo D, et al. Vestibular evoked myogenic potentials in multiple sclerosis: clinical and imaging correlations. Mult Scler 2004;10:316 321. Bandini F, Beronio A, Ghiglione E, et al. The diagnostic value of vestibular evoked myogenic potentials in multiple sclerosis. J Neurol 2004;251:617 621. Brantberg K, Granath K, Schart N. Age-related changes in vestibular evoked myogenic potentials. Audiol Neurootol 2007;12:247 253. Curthoys IS, Iwasaki S, Chihara Y, et al. The ocular vestibular-evoked myogenic potential to air-conducted sound; probable superior vestibular nerve origin. Clin Neurophysiol 2011;122:611 616. Eleftheriadou A, Deftereos SN, Zarikas V, et al. The diagnostic value of earlier and later components of vestibular evoked myogenic potentials (VEMP) in multiple sclerosis. J Vestib Res 2009;19:59 66. Jacobson GP, McCaslin DL, Piker EG, et al. Patterns of abnormality in cvemp, ovemp, and caloric tests may provide topological information about vestibular impairment. J Am Acad Audiol 2011;22:601 611. Murofushi T, Nakahara H, Yoshimura E, Tsuda Y. Association of air-conducted sound ovemp findings with cvemp and caloric test findings in patients with unilateral peripheral vestibular disorders. Acta Otolaryngol 2011;131: 945 972. Murofushi T, Shimizu K, Takegoshi H, Cheng PW. Diagnostic value of prolonged latencies in the vestibular evoked myogenic potential. Arch Otolaryngol Head Neck Surg 2001;127:1069 1072. Patkó T, Simó M, Arányi Z. Vestibular click-evoked myogenic potentials: sensitivity and factors determining abnormality in patients with multiple sclerosis. Mult Scler 2007;13:193 198. Polman CH, Reingold SC, Banwell B, et al. Diagnostic criteria for multiple sclerosis: 2010 Revisions to the McDonald criteria. Ann Neurol 2011;69:292 302. 90 Copyright Ó 2013 by the American Clinical Neurophysiology Society

Journal of Clinical Neurophysiology Volume 30, Number 1, February 2013 VEMP in MS Patients Rosengren SM, Colebatch JG. Ocular vestibular evoked myogenic potentials are abnormal in internuclear ophthalmoplegia. Clin Neurophysiol 2011;122:1264 1267. Rosengren SM, Welgampola MS, Colebatch JG. Vestibular evoked myogenic potentials: past, present and future. Clin Neurophysiol 2010;121:636 651. Sartucci F, Logi F. Vestibular-evoked myogenic potentials: a method to assess vestibulo-spinal conduction in multiple sclerosis patients. Brain Res Bull 2002;59:59 63. Shibata K, Yamane K, Otuka K, Iwata M. Abnormal visual processing in migraine with aura: a study of steady-state visual evoked potentials. J Neurol Sci 2008;271:119 126. Shimizu K, Murofushi T, Sakurai M, Halmagyi M. Vestibular evoked myogenic potentials in multiple sclerosis. J Neurol Neurosurg Psychiatry 2000;69:276 277. Versino M, Colnaghi S, Callieco R, et al. Vestibular evoked myogenic potentials in multiple sclerosis patients. Clin Neurophysiol 2002;113:1464 1469. Copyright Ó 2013 by the American Clinical Neurophysiology Society 91