Skip to main content
NCBI home page
JARO: Journal of the Association for Research in Otolaryngology logoLink to JARO: Journal of the Association for Research in Otolaryngology
. 2012 Nov 27;14(1):37–47. doi: 10.1007/s10162-012-0360-1

Frequency Tuning of the Cervical Vestibular-Evoked Myogenic Potential (cVEMP) Recorded from Multiple Sites along the Sternocleidomastoid Muscle in Normal Human Subjects

Wei Wei 1, Ben Jeffcoat 1, William Mustain 1, Hong Zhu 1, Thomas Eby 1, Wu Zhou 1,2,3,✉
PMCID: PMC3540270  PMID: 23183876

Abstract

Frequency tuning of tone burst-evoked myogenic potentials recorded from the sternocleidomastoid muscle (cervical VEMP or cVEMP) is used clinically to assess vestibular function. Understanding the characteristics of cVEMP is important for improving the specificity of cVEMP testing in diagnosing vestibular deficits. In the present study, we analyzed the frequency tuning properties of the cVEMPs by constructing detailed tuning curves and examining their morphology and dependence on SCM tonic level, sound intensity, and recording site along the SCM. Here we report two main findings. First, by employing nine tone frequencies between 125 and 4,000 Hz, some tuning curves exhibited two distinct peaks, which cannot be modeled by a single mass spring system as previously suggested. Instead, the observed tuning is better modeled as linear summation of two mass spring systems, with resonance frequencies at ~300 and ~1,000 Hz. Peak frequency of cVEMP tuning curves was not affected by SCM tonic level, sound intensity, and location of recording site on the SCM. However, sharpness of cVEMP tuning was increased at lower sound intensities. Second, polarity of cVEMP responses recorded from the lower quarter of the SCM was reversed as compared to that at the two upper sites. While more studies are needed, these results suggest that cVEMP tuning is mediated through multiple generators with different resonance frequencies. Future studies are needed to explore implications of these results on development of selective VEMP tests and determine the nature of polarity inversion at the lower quarter of SCM.

Keywords: sound activation of vestibular system, canal, otolith, VCR

Introduction

Sound-evoked vestibular responses have been well documented in humans and animal models (Bickford et al. 1964; Carey et al. 2004; Curthoys et al. 2006; McCue and Guinan 1994; Murofushi et al. 1995; Tullio 1929; Young et al. 1977; Zhou et al. 2004, 2005, 2007; Xu et. al. 2009; Zhu et al. 2011, 2012). Because sound stimulation offers simplicity, selectivity, and the capacity to stimulate each labyrinth separately, its potential value in the diagnosis of vestibular disorders has been widely recognized (Colebatch 2001, 2010; Curthoys 2010; Halmagyi et al. 2005; Minor et al. 2001; Rauch 2006; Rosengren et al. 2010; Streubel et al. 2001; Welgampola and Carey 2011; Welgampola and Colebatch 2001, 2005; Zhou and Cox 2004). Among several sound-evoked vestibular responses, the short latency myogenic potential recorded from tonically contracted sternocleidomastoid muscles (SCM) (Colebatch et al. 1994), which is known as the cVEMP, has been studied extensively and adopted by vestibular clinics worldwide. Because previous studies suggested that sound primarily activates the vestibular afferents that innervated the saccule, the cVEMP is presently used to assess the saccule function (Murofushi and Curthoys 1997; Kushiro et al. 1999; Colebatch and Rothwell 2004). However, the saccule hypothesis of sound activation of the vestibular system has been challenged by several studies (Carey et al. 2004; Curthoys and Vulovic 2011; Todd et al. 2009; Young et al. 1977; Zhou et al. 2004, 2005, 2007; Xu et al. 2009; Zhu et al. 2011). In particular, Zhu et al. (2011) recently quantitatively examined sound-evoked responses in vestibular afferents that innervated the five vestibular end organs. Their results demonstrated significant sound activation of all five end organs, although with different sensitivities.

The primary goal of the present study was to further test the saccular hypothesis of cVEMP by examining whether sources other than the saccule contribute to generating cVEMP responses. Our approach was to examine the frequency tuning of cVEMP in normal human subjects. The rationale is that frequency tuning of cVEMP reflects the physical resonance of the contributing end organ. McCue and Guinan (1994; in cats) and Zhu et al. (2012; in rats) showed that sound stimulation of the saccular afferents has a unimodal frequency response. Thus, if there in cVEMP tuning curves exhibit multiple peaks, it would indicate that the saccule is not the sole source for cVEMP and other vestibular end organs are involved. While cVEMP frequency tuning has been studied previously, the present study is different in the following three ways. First, instead of using four or six frequencies to generate cVEMP tuning curves, nine frequencies ranging from 125 to 4,000 Hz were employed in the present study. The improved frequency tuning curves revealed that cVEMP is better modeled as linear summation of two mass spring damper systems rather than a single mass spring system (Todd et al. 2000). Second, rather than recording cVEMP tuning from a single location at the upper third of the SCM, we recorded cVEMP responses at three sites along the SCM: the top quarter, the midpoint, and the lower quarter. This experiment was motivated by the possibility that similar to monkey, horse, and rabbit, which all have long SCM, the human SCM may be compartmentalized and have multiple innervation zones. We found that the cVEMP recorded at the lower quarter exhibited reversed polarity as compared to that from the two upper sites. Indeed, we found that the cVEMP at the lower quarter of the SCM exhibited opposite polarity to the cVEMPs at the other two sites. While more studies are needed to rule out alternative interpretations, these results provide new insight into developing discriminative VEMP tests and interpretative guidelines. Finally, since cVEMP response is also affected by muscle tonic contraction level, sound intensity, and recording site along the SCM (Lim et al. 1995; Li et al. 1999; Sheykholeslami et al. 2001), in the present study, we not only used nine frequencies to construct accurate tuning curves, but also examined whether these factors affect tuning curve morphology.

Methods

Nine subjects (four female, five male, from 25 to 35 years old) with informed consent participated in this study. They were free from otologic disease and were not taking any medications. The experimental protocol was approved by the Institutional Review Board at the University of Mississippi Medical Center.

During the test, subjects were instructed to sit upright and turn their heads to the right. Tonic SCM contraction was maintained by requiring subjects to push a weighted arm placed against left forehead. Two different levels of tonic contraction were produced by changing weight on the arm. Acoustic tone bursts (8 ms plateau, 1 ms rise/fall) were generated by a CED Power 1401 (Cambridge Electronics Devices, Cambridge, UK). Tone bursts at nine frequencies (125, 250, 350, 500, 750, 1,000, 1,500, 2,000, and 4,000 Hz) were presented at 124 and 118 dB SPL and tone bursts at eight frequencies (250, 350, 500, 750, 1,000, 1,500, 2,000, and 4,000 Hz) were presented at 112 dB SPL. One subject was not tested by tones at 124 dB SPL. Another subject showed no response to sound and his data were not included in our analysis. Tone bursts were randomly delivered to the left ear at a rate of 5 Hz via an insert earphone. An important feature of this stimulation delivery approach is that at each recording site, the pre-stimulus SCM tonic levels were comparable among the nine different tones (within ~5 %). Thus, the SCM tonic level did not contribute to the tuning curves. Tuning curves were measured at two SCM tonic level conditions. For example, at the low tonic level condition, the averaged rectified pre-stimulus EMG across the nine frequencies was 92 ± 3, 101 ± 3, and 92 ± 3 mV at the top, middle, and lower quarter of SCM, respectively (subject RY). At the high tonic level condition, the averaged rectified pre-stimulus EMG was 153 ± 3, 146 ± 3, and 119 ± 3 mV at the top, middle, and lower quarter of SCM, respectively, for the same subject.

It should be noted that a tuning curve was measured using the same sound intensity in the external ear canal for each tone. There was no correction for transmission of sound through the middle ear for several reasons. First, although near threshold normal hearing for low frequencies is strongly attenuated by the middle ear, at high intensities like those used in the present study, loudness contours are relatively flat below 2,000 Hz (Fletcher and Munson 1933; ISO 226:2003). Second, the middle ear transfer function expressed as displacement transmission ratio at the stapes indicates that amplitude of stapes displacement remains flat for frequencies <1,000 Hz (Gan et al. 2004). Thus, it is unclear how to make corrections for sound-induced vestibular activations. Third, for most previous VEMP tuning studies, VEMP responses were referred to pressure changes in the external canal. While we acknowledge this as an important issue to be addressed in the future, we took the current approach to be consistent with the literature. However, caution should be taken to interpret the resonance frequency data.

Surface electrodes were placed at three locations along the SCM: the top quarter, the midpoint, and the lower quarter. All active electrodes were referred to the sternum. Electrode impedance was maintained below 5 kΩ. EMG signals were amplified (2,500 gain), band path-filtered (5 to 1,000 Hz), and sampled at 10 kHz (CED Power 1401, Cambridge Electronics Devices). Data were stored on a hard disk for offline analysis.

For each frequency, cVEMP was averaged over 100 repetitions with Spike 2 (Cambridge Electronics Devices) and displayed in SigmaPlot (Systat Software Inc, CA, USA). It should be noticed that positive potentials were shown as upward deflections in this study. Amplitudes of the first (peak I) and second peak (peak II) were measured with respect to the baseline before stimulus onset. SEM is used to describe the dispersion of data.

Two models were used to fit cVEMP response at different frequencies. Equation 1 (French 1971) models frequency tuning of the cVEMP as one mass spring damper system. It was based on the hypothesis that cVEMP originated solely from the saccule. V(ω) represents the response amplitude of cVEMP (millivolt), and G is the gain of mass spring damper system. Ω is the frequency of sound stimulus, and ω0 is the resonant frequency of saccule where sound stimulus evokes the largest response. Q is the quality factor which describes the sharpness of tuning curves.

graphic file with name M1.gif 1
graphic file with name M2.gif 2

Equation 2 assumes that vestibular end organs with different resonant frequencies might be involved in the generation of cVEMP. In Eq. 2, ω1, ω2 and Q1, Q2 are the resonant frequency and quality factor for the low- and high-frequency components of cVEMP, respectively. G1 and G2 is gain of the two components, respectively. R2 was computed for each fitting and a significant component is determined if it increases R2 by >5 %. To determine the relative contributions of the low- and high-frequency components to the cVEMP, we computed a frequency selective index (FSI) at each tone frequency, which was defined as the ratio of the difference between the contributions of the high-frequency component (Ahi.f) and the low-frequency component (Alo.f) and the sum of the two components, i.e., FSIf = (Alo.f − Ahi.f) / (Ahi.f + Alo.f). Alo.f and Ahi.f are computed using the first and second items of Eq. 2, respectively. A FSIf of 1 indicates that the cVEMP at tone frequency f is contributed only by the low-frequency component. A FSIf of −1 indicates that the cVEMP at tone frequency f is contributed only by the high-frequency component. A FSIf of 0 indicates that the two components contribute equally to the cVEMP at the tone with frequency f.

cVEMP tests were also conducted in the same group of subjects with an alternate montage. In these tests, electrodes at the top, middle, and lower quarter of SCM were referenced to an electrode on the right hand. Each subject was tested by one frequency which was found to evoke the largest cVEMP response at 130 dB SPL, respectively.

Results

Figure 1 shows tone-evoked cVEMP responses (subject RY) recorded at the three sites along the SCM. While cVEMP responses at the three sites exhibited similar frequency tuning, polarity of cVEMP responses at the lower quarter site was reversed as compared to that at the two upper sites. To construct tuning curves, amplitudes of cVEMP peak I (Fig. 2A) and peak II (Fig. 2C) were plotted as a function of tone frequency. Since both tuning curves exhibited two distinct peaks, they were not adequately modeled by a single mass spring system (Eq. 1; Fig. 2A, C). However, regression lines based on Eq. 2, which models cVEMP tuning as linear summation of two resonance systems, provided better fits with increased R2 by 8 and 11 percentage points for peak I and peak II, respectively (Fig. 2B, D). The R2 values increased by 9.5 and 13.6 % for peak I and peak II tuning curves, respectively. For this subject, the peak I tuning curve was modeled as linear summation of a low-frequency system of 251 Hz and a high-frequency system of 763 Hz (R2 = 0.92), while the peak II tuning curve can be modeled as linear summation of a low-frequency system of 350 Hz and a high-frequency system of 1,035 Hz (R2 = 0.92).

FIG. 1.

FIG. 1

Open in a new tab

A–C Representative cVEMP responses (subject RY) evoked by three frequencies (125, 1,000, and 2,000 Hz) recorded at the three sites along the SCM: the lower quarter (red), the midpoint (black), and the top quarter (blue). Recordings were made at the high muscle tonic level with 124 dB SPL. Time 0 is tone onset. Upward reflection is positive and downward reflection is negative.

FIG. 2.

FIG. 2

Open in a new tab

Fitting of representative tuning curves by Eqs. 2 (B and D) and 1 (A and C) (subject RY, the top quarter site, higher tonic level, and with an intensity of 124 dB SPL).

Figures 3 (peak I) and 4 (peak II) show the tuning curves of subject RY at the three sites, three intensities at the high SCM tonic level condition. Polarity of cVEMP responses at the lower quarter of the SCM was reversed as compared to that at the other two sites in all conditions. In each tuning curve, three lines were shown to indicate the contributions of the low-frequency component (blue), the high-frequency component (purple), and the sum of the two (red). Relative contribution of the two components was assessed by computing a FSI, which is plotted as a function of frequency (the panel below each tuning curve). To illustrate individual tuning curve variability, Figure 5 shows peak I tuning curves of all the subjects at the midpoint of SCM with a sound intensity of 118 dB SPL in the high tonic level condition. Figure 6 summarizes FSI tuning curves at the three sites and at the three intensities. On average, cVEMP responses evoked by tones below 500 Hz were primarily due to activation of the low-frequency system (FSI > 0.5). The cVEMP responses evoked by tones above 1,000 Hz were primarily due to activation of the high-frequency component (FSI < −0.5).

FIG. 3.

FIG. 3

Open in a new tab

Fitting tuning curves of peak I (subject RY) as linear summation of two mass spring systems at 124 (A), 118 (B), and 112 dB (C) for each of the three recording sites. For each tuning curve, blue and purple lines are contributions of the low- and high-frequency components, respectively, and the red line is their sum. Below each tuning curve, frequency selectivity index (FSI) is plotted as a function of frequency.

FIG. 4.

FIG. 4

Open in a new tab

Fitting tuning curves of peak II (subject RY) as linear summation of two mass spring systems at 124 (A), 118 (B), and 112 dB (C) for the three recording sites. For each tuning curve, blue and purple lines are contributions of the low- and high-frequency components, respectively, and the red line is their sum. Below each tuning curve, frequency selectivity index (FSI) is plotted as a function of frequency.

FIG. 5.

FIG. 5

Open in a new tab

Tuning curves of peak I obtained from the midpoint of SCM at 118 dB at the high tonic level condition for each subject. Blue, purple, and red lines are regression lines fitted by Eq. 2 for the low-frequency component, high-frequency component, and their linear sum, respectively.

FIG. 6.

FIG. 6

Open in a new tab

Averaged frequency selectivity curves obtained from the three recording sites for three intensities. Error bars refer to SEM.

The parameters used for fitting each subject were summarized in Table 1 for 124 dB SPL. In seven out of eight subjects, their tuning curves exhibit two significant tuning peaks at all the recording sites. For the group, resonant frequency of the low-frequency component was 392 ± 42, 386 ± 35, and 333 ± 43 Hz for cVEMPs recorded at the top, midpoint, and the lower quarter of SCM, respectively. The resonant frequency of the high-frequency component was 1,240 ± 253, 1,114 ± 161, and 1,278 ± 148 Hz for cVEMP recorded at the top, midpoint, and lower quarter of SCM, respectively. There were no significant differences among the low and high resonance frequencies among the three sites (P > 0.1).

TABLE 1.

Summary of tuning curve fitting parameters at 124 dB at the high tonic level condition

Subject Top quarter Midpoint Lower quarter
f1 Q1 f2 Q2 f1 Q1 f2 Q2 f1 Q1 f2 Q2
AH 311 2.0 818 0.7 342 1.5 944 0.8 310 1.0 1,619 2.0
JG 481 0.7 N/A N/A 475 0.8 N/A N/A 495 0.9 2,034 2.5
KT 470 0.3 1,392 2.0 296 0.4 1,514 1.7 261 0.6 1,218 2.1
KA 271 1.5 653 1.5 253 1.5 699 3.0 234 1.5 976 2.0
MT 573 2.0 1,484 0.4 508 3.0 898 0.7 465 0.6 1,270 2.0
RO 465 1.4 2,694 0.8 477 1.4 1,968 1.7 327 3.0 N/A N/A
RY 251 3.0 763 1.0 431 1.0 928 3.0 426 1.0 944 3.0
WI 316 1.1 878 2.0 308 0.8 845 1.6 150 1.0 884 0.9
Mean 392 1.5 1,240 1.2 386 1.3 1,114 1.8 333 1.2 1,278 2.1
SEM 42 0.3 253 0.2 35 0.3 161 0.3 43 0.3 148 0.2
Open in a new tab

For each subject, f1 and f2 stand for the resonant frequency of the low- and high-frequency components, respectively; Q1 and Q2 stand for the quality factor of the low- and high-frequency components, respectively. Top quarter, midpoint, and lower quarter refer to the three sites we record cVEMP over the SCM

N/A not applicable

The effects of recording site, SCM tonic level, and tone intensity are summarized in Figures 7 and 8. First, ANOVA analysis shows that SCM tonic level has no effect on f1, f2, Q1, and Q2 (P  > 0.05, data not shown). Thus, data obtained from the two tonic levels were combined in later analysis. As shown in Figure 7 (peak I), f1 and f2 were neither affected by recording site (f1: P = 0.185; f2: P = 0.985) nor by stimulus intensity (f1: P = 0.066; f2: P = 0.813). For the tuning curves of cVEMP peak I recorded at the top quarter, the midpoint, and the lower quarter of SCM, f1 was 418 ± 18, 416 ± 18, and 373 ± 20 Hz, respectively, and f2 was 1,205 ± 73, 1,188 ± 71, and 1,192 ± 80 Hz, respectively. Similarly, for the tuning curves of cVEMP peak I recorded at 124, 118, and 112 dB, f1 was 371 ± 19, 403 ± 19, and 433 ± 19 Hz, respectively, and f2 was 1,188 ± 72, 1,232 ± 73, and 1,165 ± 79 Hz, respectively. Neither Q1 nor Q2 was affected by recording site (Q1: P = 0.258; Q2: P = 0.704). Stimulus intensity has a significant effect on Q1 (P < 0.001), but not on Q2 (P = 0.087). Q1 at 124, 118, and 112 dB was 1.4 ± 0.1, 1.7 ± 0.1, and 2.3 ± 0.1, respectively. There is no interaction between stimulus intensity and recording site for f1, f2, Q1, and Q2 (P > 0.474).

FIG. 7.

FIG. 7

Open in a new tab

A–D Effects of stimulus intensity and recording site on frequency tuning of peak I. Error bars refer to SEM. Statistical significance (P < 0.05) is indicated by asterisk. For details, see text.

FIG. 8.

FIG. 8

Open in a new tab

A–D Effects of stimulus intensity and recording site on frequency tuning of peak II. Error bars refer to SEM. Statistical significance (P < 0.05) is indicated by asterisk. For details, see text.

Similar results were obtained for peak II (Fig. 8). Recording site has no effect on peak frequency (f1: P = 0.967; f2: P = 0.948) and quality factor (Q1: P = 0.106; Q2: P = 0.844). Stimulus intensity has no significant effect on f1 or f2 (f1: P = 0.252; f2: P = 0.102), but a decrease in intensity significantly increases quality factor (Q1: P = 0.001; Q2: P = 0.003). At 124, 118, and 112 dB, Q1 was 1.3 ± 0.1, 1.4 ± 0.1, and 1.8 ± 0.1, respectively, and Q2 was 1.6 ± 0.1, 1.8 ± 0.1, and 2.1 ± 0.1, respectively. There is no interaction between stimulus intensity and recording site for f1, f2, Q1, and Q2 (P  > 0.254).

The effects of recording site and intensity on latency and duration are summarized in Figure 9. Recording site had a significant effect on response latency and duration (P < 0.001), but tone intensity had no effect on response latency (P = 0.09) and duration (P = 0.677). There is no interaction between stimulus intensity and recording site for latency (P = 0.07) and duration (P = 0.06). Latency of peak I from the lower quarter recording site was shorter than peak I recorded from the midpoint (14.7 ± 0.2 vs. 17.3 ± 0.1 ms, P < 0.001) and the top quarter (14.7 ± 0.2 vs. 16.3 ± 0.2 ms, P < 0.001). Latency of peak I from the midpoint recording site was shorter than that of the top quarter site (P < 0.001). Peak II latency from the lower quarter site was shorter than peak II from the midpoint site (23.8 ± 0.2 vs. 28.9 ± 0.2 ms, P < 0.001) and the top quarter site (23.8 ± 0.2 vs. 28.4 ± 0.2 ms, P < 0.001), but latency of peak II from the top quarter site was not significantly different from peak II recorded at the midpoint (P = 0.114). Duration of peak I–peak II recorded at the lower quarter was reduced as compared to the response duration at the midpoint (9.1 ± 0.2 vs. 11.6 ± 0.2 ms, P < 0.001) and the top quarter (9.1 ± 0.2 vs. 12.1 ± 0.2 ms, P < 0.001). The response duration at the midpoint recording site was also shorter than that of the top quarter site (P = 0.014).

FIG. 9.

FIG. 9

Open in a new tab

Effects of recording site on latency of peak I (A) and peak II (B) and the duration (C) between the two peaks. Error bars refer to SEM. Statistical significance (P < 0.05) is indicated by asterisk.

Adoption of wrist reference has no effect on the polarity of cVEMP at the top, middle, and lower quarter of SCM. Compared to the positive–negative waveform of cVEMP recorded from the upper two sites of SCM, the polarity of cVEMP found at the lower quarter site remained to be reversed in all the subjects tested.

Discussion

In this study, we extended cVEMP frequency tuning analysis by constructing more detailed tuning curves, examining their morphology and characterizing their dependence on sound sensitivity, recording site, and SCM tonic level in normal human subjects. There are two main findings in this study. First, we found evidence that cVEMP tuning is better modeled as linear summation of two mass spring systems. Second, cVEMP responses recorded at the lower quarter exhibited reversed polarity as compared to that recorded at the two upper sites.

Since the properties of cVEMP tuning are related to the physical properties of the vestibular end organs that are activated by sound, cVEMP tuning analysis has been employed to characterize the putative cVEMP peripheral generators. Motivated by the hypothesis that sound primarily activates the saccule (Curthoys et al. 2006; Murofushi et al. 1995; Murofushi and Curthoys 1997), Todd et al. (2000) modeled cVEMP tuning as a single mass spring system and reported a resonant frequency of ~300 Hz. However, later studies showed that cVEMP tuning curves were characterized by a dominant peak between 400 and 800 Hz (Akin et al. 2003; Janky and Shepard 2009; Murofushi et al. 1999; Rauch et al. 2004; Timmer et al. 2006; Todd et al. 2009; Welgampola and Colebatch 2001). In the present study, cVEMP tuning curves were constructed with more tone frequencies between 125 and 4,000 Hz. Since four out of nine detailed tuning curves showed two distinct peaks, they were better modeled as linear summation of two mass spring systems. Our results show that cVEMPs are generated by activation of two components: one resonating at ~300 Hz and the other at ~1,000 Hz. The low-frequency component accounts for more than 75 % (FSI > 0.5) of the overall response for tones below 500 Hz, whereas the high-frequency component makes up more than 75 % (FSI < −0.5) of the overall response for tones above 1,000 Hz. Similar double tuning peaks were in a recent study (Zhang et al. 2012), which found a tuning peak around 100 Hz in addition to a dominant peak around 600 Hz. The 100-Hz peak was considered to be of utricle origin and the 600-Hz peak of saccular origin. Compared to the study of Zhang et al., our tuning peaks are located at higher frequencies. This difference might be due to the different frequency ranges chosen by the two studies (50–1,200 vs. 125–4,000 Hz). Nevertheless, these new results are consistent with accumulating evidence from animal model studies indicating that the saccule is not the sole generator of sound-evoked vestibular responses (Carey et al. 2004; Curthoys and Vulovic 2011; Wit et al. 1984; Young et al. 1977; Zhou et al. 2004, 2005, 2007; Xu et al. 2009). In particular, a recent study by Zhu et al. (2011) showed that 73 % irregular utricle afferents as well as 43 % irregular canal afferents were sensitive to clicks. This multiple generator hypothesis of sound-evoked vestibular responses is also consistent with our previous studies on frequency tuning of bone- and air-conducted VEMP recorded from extraocular muscles (oVEMP), which showed that the tuning peak of air-conducted oVEMP (~1,000 Hz) was much higher than the tuning peak of bone-conducted oVEMP (~400 Hz) (Donnellan et al. 2010; Lewis et al. 2010). We hypothesize that utricle and canals also contribute to cVEMP responses. However, it is premature to associate the low-frequency component and the high-frequency component in our study to a specific vestibular end organ. Additional studies in humans to further explore this hypothesis may provide important insight for improving clinical VEMP tests.

In addition to characterize tuning properties of the cVEMP, we found that the cVEMP responses recorded at the lower quarter site exhibited a negative–positive waveform, rather than the typical positive–negative waveform observed in the two upper sites. There are two interpretations for the polarity reversal in cVEMPs recorded at the lower quarter of the SCM. One interpretation is that the polarity reversal is resulted from volume conduction of motor unit action potentials. Indeed, it has been shown that volume conduction can lead to polarity inversion near the muscle–tendon junction while the polarity of waveforms simultaneously recorded from the middle and top portions of muscle remains unchanged (Roeleveld et al. 1997; Dumitru and King 1991). The other interpretation is that the SCM is compartmentalized, and there exist multiple innervation zones (McLoon 1998). Previous studies showed that the endplates in horse SCM and rabbit SCM were organized into multiple bands 1–3 cm apart throughout the length of muscle (Zenker et al. 1990; McLoon 1998). Given the length of human SCM muscle, it is possible that human SCM is also compartmentalized. However, it has not been shown in published reports. Furthermore, as initial positivity of cVEMP responses reflects sound-induced inhibitory modulation of ipsilateral SCM motoneurons, initial negativity of cVEMP responses reflects sound-induced excitatory modulation of ipsilateral SCM motoneurons (Colebatch and Rothwell 2010). Although Uchino et al. found in cats that some IPSPs evoked by electrical stimulation of saccule nerve were preceded by an EPSP (Kushiro et al. 1999), it remains to be determined whether human SCM receives similar excitatory input from the ipsilateral vestibular pathways. Therefore, it is an open question on how the polarity reversal is generated, and future studies are needed to solve the nature of polarity reversal observed in the present study. It is also worth noting that the possibility of having differential nerve innervation zones along the SCM is not directly related to the hypothesis of multiple generators of cVEMP because a single end organ (e.g., the saccule) may contribute to all parts of the SCM but with different connectivity patterns.

Acknowledgments

We thank Jerome Allison and Ron Ron Cheng for their help during the project.

Contributor Information

Wei Wei, Email: weiwe@musc.edu.

Ben Jeffcoat, Email: bjeffcoat@mbhs.org.

William Mustain, Email: WMustain@umc.edu.

Hong Zhu, Email: HOZhu@umc.edu.

Thomas Eby, Email: TEby@umc.edu.

Wu Zhou, Phone: +1-601-8154735, FAX: +1-601-9841655, Email: wzhou@umc.edu.

References

  1. Akin FW, Murnane OD, Proffitt TM. The effects of click and tone-burst stimulus parameters on the vestibular evoked myogenic potential (VEMP) J Am Acad Audiol. 2003;14:500–509. doi: 10.3766/jaaa.14.9.5. [DOI] [PubMed] [Google Scholar]
  2. Bickford RG, Jacobson JL, Cody DTR. Nature of average evoked potentials to sound and other stimuli in man. Ann NY Acad Sci. 1964;112:204–18. doi: 10.1111/j.1749-6632.1964.tb26749.x. [DOI] [PubMed] [Google Scholar]
  3. Carey JP, Hirvonen TP, Hullar TE, Minor LB. Acoustic responses of vestibular afferents in a model of superior canal dehiscence. Otol Neurotol. 2004;25:345–352. doi: 10.1097/00129492-200405000-00024. [DOI] [PubMed] [Google Scholar]
  4. Colebatch JG. Vestibular evoked potentials. Curr Opin Neurol. 2001;14:21–26. doi: 10.1097/00019052-200102000-00004. [DOI] [PubMed] [Google Scholar]
  5. Colebatch JG. Sound conclusion? Clin Neurophysiol. 2010;121:124–126. doi: 10.1016/j.clinph.2009.09.026. [DOI] [PubMed] [Google Scholar]
  6. Colebatch JG, Halmagyi GM, Skuse NF. Myogenic potentials generated by a click-evoked vestibulo-collic reflex. J Neurol Neurosurg Psychiatry. 1994;57:190–197. doi: 10.1136/jnnp.57.2.190. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Colebatch JG, Rothwell JC. Motor unit excitability changes mediating vestibulocollic reflexes in the sternocleidomastoid muscle. Clin Neurophysiol. 2004;115(11):2567–2573. doi: 10.1016/j.clinph.2004.06.012. [DOI] [PubMed] [Google Scholar]
  8. Curthoys IS, Kim J, McPhedran SK, Camp AJ. Bone conducted vibration selectively activates irregular primary otolithic vestibular neurons in the guinea pig. Exp Brain Res. 2006;175:256–267. doi: 10.1007/s00221-006-0544-1. [DOI] [PubMed] [Google Scholar]
  9. Curthoys A critical review of the neurophysiological evidence underlying clinical vestibular testing using sound, vibration and galvanic stimuli. Clin Neurophysiol. 2010;121(2):132–44. doi: 10.1016/j.clinph.2009.09.027. [DOI] [PubMed] [Google Scholar]
  10. Curthoys IS, Vulovic V. Vestibular primary afferent responses to sound and vibration in the guinea pig. Exp Brain Res. 2011;210:347–352. doi: 10.1007/s00221-010-2499-5. [DOI] [PubMed] [Google Scholar]
  11. Donnellan K, Wei W, Jeffcoat B, Mustain W, Xu Y, Eby T, Zhou W. Frequency tuning of bone-conducted tone burst-evoked myogenic potentials recorded from extraocular muscles (BOVEMP) in normal human subjects. Laryngoscope. 2010;120:2555–2560. doi: 10.1002/lary.21100. [DOI] [PubMed] [Google Scholar]
  12. Dumitru D, King JC (1991) Far-field potentials in muscle. Muscle nerve 14:981–989 [DOI] [PubMed]
  13. Fletcher H, Munson W. Loudness, its definition, measurement and calculation. J Acoust Soc Am. 1933;5:82–108. doi: 10.1121/1.1915637. [DOI] [Google Scholar]
  14. French AP. Vibrations and waves. New York: W. W. Norton; 1971. [Google Scholar]
  15. Gan RZ, Wood MW, Dormer KJ. Human middle ear transfer function measured by double laser interferometry system. Otol Neurotol. 2004;25:423–435. doi: 10.1097/00129492-200407000-00005. [DOI] [PubMed] [Google Scholar]
  16. Halmagyi GM, Curthoys IS, Colebatch JG, Aw ST. Vestibular responses to sound. Ann N Y Acad Sci. 2005;1039:54–67. doi: 10.1196/annals.1325.006. [DOI] [PubMed] [Google Scholar]
  17. Janky KL, Shepard N. Vestibular evoked myogenic potential (VEMP) testing: normative threshold response curves and effects of age. J Am Acad Audiol. 2009;20(8):514–522. doi: 10.3766/jaaa.20.8.6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kushiro K, Zakir M, Ogawa Y, Sato H, Uchino Y. Saccular and utricular inputs to sternocleidomastoid motoneurons of decerebrate cats. Exp Brain Res. 1999;126:410–416. doi: 10.1007/s002210050747. [DOI] [PubMed] [Google Scholar]
  19. Lewis AF, Mustain W, Xu Y, Eby T, Zhou W. Frequency tuning in the tone burst-evoked myogenic potentials in extraocular muscles in normal human subjects. J Otolaryngol Head Neck Surg. 2010;39:491–497. [PubMed] [Google Scholar]
  20. Li MW, Houlden D, Tomlinson RD. Click evoked responses in sternocleidomastoid muscles: characteristics in normal subjects. J Vestib Res. 1999;9:327–334. [PubMed] [Google Scholar]
  21. Lim CL, Clouston P, Sheean G, Yiannikas C. The influence of voluntary EMG activity and click intensity on the vestibular click evoked myogenic potential. Muscle Nerve. 1995;18(10):1210–1213. doi: 10.1002/mus.880181021. [DOI] [PubMed] [Google Scholar]
  22. McCue MP, Guinan JJ. Acoustically responsive fibers in the vestibular nerve of the cat. J Neurosci. 1994;14:6058–6070. doi: 10.1523/JNEUROSCI.14-10-06058.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. McLoon LK. Muscle fiber type compartmentalization and expression of an immature myosin isoform in the sternocleidomastoid muscle of rabbits and primates. J Neurol Sci. 1998;156:3–11. doi: 10.1016/S0022-510X(98)00011-2. [DOI] [PubMed] [Google Scholar]
  24. Minor LB, Cremer PD, Carey JP, Della Santina CC, Streubel SO, Weg N. Symptoms and signs in superior canal dehiscence syndrome. Ann NY Acad Sci. 2001;942:259–273. doi: 10.1111/j.1749-6632.2001.tb03751.x. [DOI] [PubMed] [Google Scholar]
  25. Murofushi T, Curthoys IS, Topple AN, Colebatch JG, Halmagyi GM. Responses of guinea pig primary vestibular neurons to clicks. Exp Brain Res. 1995;103:174–178. doi: 10.1007/BF00241975. [DOI] [PubMed] [Google Scholar]
  26. Murofushi T, Curthoys IS. Physiological and anatomical study of click sensitive primary vestibular afferents in the guinea pig. Acta Oto-Laryngol. 1997;117:66–72. doi: 10.3109/00016489709117994. [DOI] [PubMed] [Google Scholar]
  27. Murofushi T, Matsuzaki M, Wu CH (1999) Short tone burst-evoked myogenic potentials on the sternocleidomastoid muscle. Arch Otolaryngol Head Neck Surg 125:660–664 [DOI] [PubMed]
  28. Rauch SD. Vestibular evoked myogenic potentials. Current Opinion in Otolaryngology & Head and Neck Surgery. 2006;14:299–304. doi: 10.1097/01.moo.0000244185.65022.01. [DOI] [PubMed] [Google Scholar]
  29. Rauch SD, Zhou G, Kujawa SG, et al. Vestibular evoked myogenic potentials show altered tuning in patients with Meniere’s disease. Otol Neurotol. 2004;25:333–338. doi: 10.1097/00129492-200405000-00022. [DOI] [PubMed] [Google Scholar]
  30. Roeleveld K, Blok JH, Stegeman DF, van Oosterom A. Volume conduction models for surface EMG; confrontation with measurements. J Electromyogr Kinesiol. 1997;7(4):221–232. doi: 10.1016/S1050-6411(97)00009-6. [DOI] [PubMed] [Google Scholar]
  31. Rosengren SM, Welgampola MS, Colebatch JG. Vestibular evoked myogenic potentials: past, present and future. Clin Neurophysiol. 2010;121:636–651. doi: 10.1016/j.clinph.2009.10.016. [DOI] [PubMed] [Google Scholar]
  32. Sheykholeslami K, Murofushi T, Kaga K. The effect of sternocleidomastoid electrode location on vestibular evoked myogenic potential. Auris Nasus Larynx. 2001;28(1):41–3. doi: 10.1016/S0385-8146(00)00091-2. [DOI] [PubMed] [Google Scholar]
  33. Streubel SO, Cremer PD, Carey JP, Weg N, Minor LB. Vestibular-evoked myogenic potentials in the diagnosis of superior canal dehiscence syndrome. Acta Otolaryngol Suppl. 2001;545:41–49. doi: 10.1080/000164801750388090. [DOI] [PubMed] [Google Scholar]
  34. Timmer FCA, Zhou G, Guinan JJ, Kujawa SG, Herrmann BS, Rauch SD. Vestibular evoked myogenic potentials (VEMP) in patients with Meniere’s disease with drop attacks. Laryngoscope. 2006;116:776–779. doi: 10.1097/01.mlg.0000205129.78600.27. [DOI] [PubMed] [Google Scholar]
  35. Todd NP, Cody FW, Banks JR. A saccular origin of frequency tuning in myogenic vestibular evoked potentials? Implications for human responses to loud sounds. Hear Res. 2000;141:180–8. doi: 10.1016/S0378-5955(99)00222-1. [DOI] [PubMed] [Google Scholar]
  36. Todd NPM, Rosengren SM, Colebatch JG. A utricular origin of frequency tuning to low frequency vibration in the human vestibular system. Neurosci Lett. 2009;451:175–80. doi: 10.1016/j.neulet.2008.12.055. [DOI] [PubMed] [Google Scholar]
  37. Tullio P. Das Ohr und die Entstehung der Sprache und Schrift. Berlin: Urban & Schwarzenberg; 1929. [Google Scholar]
  38. Welgampola MS, Carey JP. Waiting for the evidence: VEMP testing and the ability to differentiate utricular vs. saccular function. Otolaryngol Head Neck Surg. 2011;143:281–283. doi: 10.1016/j.otohns.2010.05.024. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Welgampola MS, Colebatch JG. Vestibulocollic reflexes: normal values and the effect of age. Clin Neurophysiol. 2001;112:1971–1979. doi: 10.1016/S1388-2457(01)00645-9. [DOI] [PubMed] [Google Scholar]
  40. Welgampola MS, Colebatch JG. Characteristics and clinical applications of vestibular-evoked myogenic potentials. Neurology. 2005;64:1682–1688. doi: 10.1212/01.WNL.0000161876.20552.AA. [DOI] [PubMed] [Google Scholar]
  41. Wit HP, Bleeker JD, Mulder HH. Responses of pigeon vestibular nerve fibers to sound and vibration with audiofrequencies. J Acoust Soc Am. 1984;75(1):202–208. doi: 10.1121/1.390396. [DOI] [PubMed] [Google Scholar]
  42. Xu Y, Simpson I, Tang X, Zhou W. Acoustic clicks activate both the canal and otolith vestibulo-ocular reflex pathways in behaving monkeys. JARO. 2009;10(4):569–77. doi: 10.1007/s10162-009-0178-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Young ED, Fernandez C, Goldberg JM. Responses of squirrel monkey vestibular neurons to audio-frequency sound and head vibration. Acta Otolaryngol. 1977;84:352–360. doi: 10.3109/00016487709123977. [DOI] [PubMed] [Google Scholar]
  44. Zenker W, Snobl D, Boetschi R. Multifocal innervation and muscle length. A morphological study on the role of myomyonal junctions, fiber branching and multiple innervation in muscles of different size and shape. Anat Embryol. 1990;182:273–283. doi: 10.1007/BF00185520. [DOI] [PubMed] [Google Scholar]
  45. Zhang AS, Govender S, Colebatch JG. Tuning of the ocular vestibular evoked myogenic potential (oVEMP) to AC sound shows two separate peaks. Exp Brain Res. 2012;213(1):111–6. doi: 10.1007/s00221-011-2783-z. [DOI] [PubMed] [Google Scholar]
  46. Zhou G, Cox LC. Vestibular evoked myogenic potentials: history and overview. Am J Audiol. 2004;13:135–143. doi: 10.1044/1059-0889(2004/018). [DOI] [PubMed] [Google Scholar]
  47. Zhou W, Mustain W, Simpson I. Sound-evoked vestibulo-ocular reflexes (VOR) in trained monkeys. Exp Brain Res. 2004;156:129–134. doi: 10.1007/s00221-003-1778-9. [DOI] [PubMed] [Google Scholar]
  48. Zhou W, Simpson I, Xu Y, Fong A. Activity-dependent modulation: a non-linearity in the unilateral vestibulo-ocular reflex (VOR) pathways. Exp Brain Res. 2005;163:267–272. doi: 10.1007/s00221-005-2304-z. [DOI] [PubMed] [Google Scholar]
  49. Zhou W, Xu Y, Simpson I, Cai YD. Multiplicative computation in the vestibulo-ocular reflex (VOR) J Neurophysiol. 2007;97:2780–2789. doi: 10.1152/jn.00812.2006. [DOI] [PubMed] [Google Scholar]
  50. Zhu H, Tang X, Wei W, Mustain W, Xu Y, Zhou W. Click-evoked responses in vestibular afferents in rats. J Neurophysiol. 2011;106(2):754–63. doi: 10.1152/jn.00003.2011. [DOI] [PubMed] [Google Scholar]
  51. Zhu H, Tang X, Wei W, Mustain W, Wood P, Xu Y, Zhou W (2012). Air-conducted short tone bursts-evoked vestibular responses in rats. Annual Meeting of Society for Neuroscience

Articles from JARO: Journal of the Association for Research in Otolaryngology are provided here courtesy of Association for Research in Otolaryngology

RESOURCES