February 10, 2009 Features

Clinical Assessment of Otolith Function

The vestibular system comprises two types of sensory organs—the semicircular canals and the otolith organs. The three semicircular canals (superior, posterior, and horizontal) are positioned orthogonally to one another to sense rotational head movement or angular acceleration in the yaw, pitch, and roll planes. The primary role of the semicircular canals is to provide input to the vestibulo-ocular reflex for gaze stability when the head rotates.

The two otolith organs (the saccule and utricle) are positioned perpendicular to each other and sense linear acceleration, head tilt, and gravity, with the primary role of providing input to the vestibulospinal reflex for postural stability. The vestibulospinal reflex serves to modulate posture via two descending pathways that aid in tonic contractions of the antigravity muscles in the arms and legs (lateral vestibulospinal tract) and activate neck motoneurons for the coordination of neck and eye movements (medial vestibulospinal tract). The lateral vestibulospinal tract receives the majority of its input from the otoliths and the cerebellum, whereas the medial vestibulospinal tract receives the majority of its input from the semicircular canals.

It is well-established that the vestibular system, and the saccule in particular, is sensitive to sound. In some lower vertebrates such as amphibians and fish, the saccule is the organ of hearing (Popper et al., 1982; Moffat & Capranica, 1976). Although the cochlea has evolved to replace the saccule as the primary organ of hearing in mammals, evidence indicates that the mammalian saccule and vestibular afferent nerve fibers are responsive to sound (e.g., McCue & Guinan, 1995).

Supplementing the Conventional Vestibular Test Battery

Conventional vestibular assessment using the electronystagmography (ENG) test battery has been limited to the evaluation of the horizontal semicircular canal and the superior branch of the vestibular nerve. Traditionally, normal vestibular function is inferred from normal ENG findings. Audiovestibular pathologies, however, can affect either branch of the vestibular nerve and any of the five peripheral vestibular end organs.

Specifically, vestibular disorders affecting the superior and posterior canals and/or the otolith organs may not be detected using a traditional vestibular test battery. Stockwell (2000) reported that 61% of patients complaining of dizziness and/or imbalance had normal ENG examinations (i.e., normal horizontal semicircular canal and superior vestibular nerve function). A more comprehensive vestibular test battery may identify vestibular involvement undetected by tests of horizontal semicircular canal function alone.

Until recently, tests to measure otolith function have been used experimentally rather than clinically. Experimental methods include the measurement of the otolith-ocular response during stimulation of the otoliths with linear acceleration using parallel sleds and swings. Unfortunately, these methods of otolith stimulation are cumbersome and not sensitive to unilateral otolith hypofunction, limiting their use to experimental settings.

Recent research has focused on the vestibular-evoked myogenic potential (VEMP) and the subjective visual-vertical (SVV) tests as measures of otolith function. VEMP and SVV supplement the vestibular test battery by providing information about otolith function to improve the differential diagnosis of patients with balance disorders. Specifically, the VEMP reflects saccular function and the SVV has been used as a test of utricular function.

Vestibular-Evoked Myogenic Potential

The VEMP is established as a clinical test of saccular (one of the otolith organs) and inferior vestibular nerve function (Colebatch, 2001). VEMPs are short-latency electromyograms evoked by high-level acoustic stimuli recorded from surface electrodes over the tonically contracted sternocleidomastoid (SCM) muscle.

In our laboratory, VEMPs are recorded with the patient seated upright and the head turned to one side (away from the stimulus ear) to activate unilaterally the SCM muscle. The SCM muscle can be activated bilaterally by raising the head from supine. A two-channel recording of the VEMP is obtained using a commercially available evoked-potential unit. Non-inverting electrodes are placed at the midpoint of the SCM muscle on each side of the neck, the inverting electrodes are placed at the sternoclavicular junctions, and the ground electrode is placed on the forehead. Clicks or tone bursts are presented monaurally at stimulus levels that range from 90 to100 dB nHL.

The response characteristics used to describe and interpret VEMP waveforms include P1 and N1 latency, P1-N1 amplitude, threshold, and asymmetry ratio. In our laboratory, P1-N1 amplitudes range from 13–178 µV for clicks (100 dB nHL) and from 15–337 µV for 500-Hz tone bursts (90 dB nHL). VEMP thresholds for clicks range from 80 to 100 dB nHL and from 75 to 85 dB nHL for 500 Hz tone bursts. Because of the large inter-subject variability for amplitude, the clinical interpretation of the VEMP has focused primarily on the asymmetry ratio. Asymmetry ratios range from 0% to ~40% in healthy individuals (e.g., Li et al., 1999).

The VEMP is dependent on normal vestibular function as it is abolished after vestibular nerve section. However, the VEMP is independent of cochlear function as it is preserved in patients with severe-to-profound sensorineural hearing loss (Colebatch & Halmagyi, 1992). In addition to normal vestibular function, the VEMP's presence is dependent upon adequate acoustic stimulation and ipsilateral activation of the SCM muscle. The neurophysiological and clinical data indicate that VEMPs are mediated by an ipsilateral pathway that includes the saccular macula, the inferior vestibular nerve, the vestibular nucleus, the vestibulospinal tract, the spinal accessory nerve, and the SCM muscle (e.g., Colebatch & Halmagyi, 1992; McCue & Guinan, 1995; Basta et al., 2005).

Influence of Non-Pathological Factors on the VEMP

Several stimulus and recording parameters affect the VEMP. These include stimulus level, stimulus frequency, the level of SCM muscle activation, the mode of stimulus presentation (air conduction versus bone conduction), and patient age.

  • Stimulus Effects. VEMP amplitude increases as a function of the stimulus level, whereas VEMP latency is independent of the stimulus level (e.g., Akin et al., 2003). The maximum VEMP amplitudes for air conduction stimuli are obtained for stimulus frequencies from 500 to 1000 Hz and the amplitude-frequency functions are consistent with the neurophysiological finding that acoustically responsive afferent fibers in the mammalian inferior vestibular nerve have best frequencies between 500 and 1000 Hz (McCue & Guinan, 1995). In contrast, stimulus frequency has little or no effect on VEMP latency if the stimulus rise/fall time remains constant. In individuals with normal audiovestibular function, VEMP air-conduction thresholds range from 100 to 120 dB peakSPL with the lowest (best) thresholds obtained at 500 and 750 Hz, the highest thresholds obtained at 2000 Hz, and no responses obtained above 2000 Hz (Akin et al., 2003).
  • EMG Effect. The tonic state of the SCM muscle is a vital factor in the recording method of the VEMP because the P1-N1 amplitude is proportional to the level of SCM muscle activity (e.g., Akin et al., 2004). Because clinical interpretation of the VEMP typically is based on inter-ear amplitude comparisons, differences in muscle activity levels between the right and left sides must be either measured or monitored. Several methods have been used to control or monitor SCM muscle activity (Colebatch et al., 1994; Akin & Murnane, 2001; Vanspauwen, Wuyts, & Van de Heyning, 2006).
  • Bone Conduction Effects. VEMPs can be elicited by acoustic stimuli presented via air conduction and bone conduction. Air conduction VEMPs require a normal middle-ear conductive mechanism to convey the stimulus to the vestibular end organs and are typically reduced or absent in patients with conductive hearing loss, whereas bone conduction VEMPs are typically present in these patients (e.g., Sheykholeslami et al., 2000). There are notable differences in the VEMP response characteristics for bone conduction versus air conduction stimuli. Specifically, the maximum amplitudes for bone conduction VEMPS are obtained at lower frequencies (typically 200–50 Hz) compared to air conduction VEMPs, and bone conduction VEMP thresholds are 40–50 dB lower than air conduction VEMP thresholds. The lower thresholds for bone-conduction VEMPs compared to air-conduction VEMPs may be related to the activation of both utricular and saccular afferents when using bone conduction stimuli (Curthoys et al., 2006). Bone-conduction VEMPs are a useful method to assess saccular and/or inferior vestibular nerve function in patients with conductive hearing loss or in patients for whom the use of high-level air conduction stimuli may be contraindicated (e.g., patients with hyperacusis). In addition to acoustic stimuli, VEMPs have been elicited using skull taps and galvanic stimulation.
  • Age Effects. Age-related anatomical and physiological degenerative changes in the vestibular system are well-documented. Several studies show a decrease in VEMP amplitude in individuals older than 60, indicating that aging affects human saccular and/or inferior vestibular nerve function (e.g., Welgampola & Colebatch, 2001). Akin and Murnane (2008) demonstrated that the decrement in VEMP amplitude due to aging was independent of the effects of aging on SCM muscle activity.

Clinical Application of the VEMP

The diagnostic utility of the VEMP has been examined for various audiovestibular and neurological disorders including vestibular neuritis, Ménière's disease, superior semicircular canal dehiscence, large vestibular aqueduct syndrome, the Tullio phenomenon, vestibular schwannoma, conductive hearing loss, sensorineural hearing loss, multiple sclerosis, spinocerebellar degeneration, and brainstem lesions.

VEMP abnormalities vary across pathologies; however, in general, inter-ear amplitude differences and an absent response are the most common abnormalities in vestibular-related disorders. The main exception to this rule is the abnormally low VEMP threshold that occurs in patients with superior semicircular canal dehiscence, large vestibular aqueduct syndrome, and Tullio phenomenon. Prolonged latency is most common in patients with central pathologies such as multiple sclerosis.

Recent Development: Ocular VEMP

Recently, Rosengren, Todd, and Colebatch (2005) described an ocular VEMP recorded from the extra-ocular muscles by activating the otoliths with an acoustic stimulus. In contrast to the VEMP recorded from the SCM muscle (cervical VEMP), the ocular VEMP reflects bilateral but predominantly contralateral otolith-ocular function and requires only that the patient sit quietly and fix his or her gaze on a stationary visual target. The normal response characteristics of the ocular VEMP are similar to those of the cervical VEMP in terms of threshold, frequency response, and level of muscle activation. The ocular VEMP may complement the cervical VEMP by providing a comprehensive evaluation of the vestibular pathways to the extraocular muscles as well as to the cervical spinal cord. The ocular VEMP may also be used as an alternative method for measuring saccular function in patients for whom the cervical VEMP cannot be recorded due to inadequate contraction of the SCM muscle.

Subjective Visual Vertical Test (SVV)

The otoliths act as gravito-inertial force (GIF) sensors and contribute to the perception of spatial orientation (earth verticality). The SVV is a psychophysical measure of the angle between perceptual vertical and true (gravitational) vertical that has been used as a clinical test of utricular function. The SVV can be measured with the patient in the static position (stationary and upright), during rotation around the vertical axis (on-axis), and during rotation around the vertical axis with the patient off-set from the center (off-axis).

Static SVV. In the upright, static position, normal individuals align a visual linear marker within 2° of true vertical (0˚). Impaired static SVV has been documented in patients with acute unilateral vestibular disorders and in patients with central vestibular disorders such as brainstem and thalamic infarctions. During acute vestibulopathy, patients may tilt the static SVV up to 21° toward the affected side (Bohmer & Rickenmann, 1995). The offset of the SVV may be due to ocular torsion (a torsional deviation of the eyes). Although the mechanism of ocular torsion is not fully understood, it is part of the ocular tilt reaction of the otolith system and considered analogous to spontaneous nystagmus of the semicircular canals. Once vestibular compensation or recovery is complete, static SVV returns to normal (±2˚). The static SVV, therefore, has been used clinically as a measure of vestibular compensation.

On-Axis Rotation. Because the utricles are positioned 3.5–4 cm from midline, rotating an individual around the vertical axis (on-axis) at constant velocity also stimulates the otolith organs. Specifically, constant on-axis rotation creates a centrifugal force (linear acceleration) that activates the utricular maculae (sensory cells of the utricle) on both sides of the head. It is presumed that the utricle is stimulated rather than the saccule because the linear acceleration along the inter-aural axis corresponds to the primary axis of the utricular maculae. Both utricles are exposed to equal and opposite centrifugal force (equal GIF) resulting in cancellation of the stimulus (and no perception of tilt) to each utricle in normal individuals. If one otolith is hypofunctional, however, then the on-axis SVV is tilted toward the diseased side.

Off-Axis Rotation. The SVV also can be measured during rotation around the vertical axis with the patient offset from the center (called off-axis rotation or unilateral centrifugation). Unilateral centrifugation measures ear-specific utricular function and determines chronic (compensated) vestibular dysfunction.

To perform this test, an individual is rotated at a constant velocity (240°–400°/s) with the test ear typically positioned 7–8 cm off-axis and the non-test ear positioned on-axis. During constant velocity rotation, the vestibulo-ocular response of the horizontal semicircular canals diminishes, and the off-axis rotation creates a centrifugal force (or linear acceleration) that stimulates the utricle positioned off-axis. In contrast to on-axis rotation (bilateral centrifugation), only the test ear (placed off-axis) is stimulated during unilateral centrifugation resulting in increased GIF to the test ear.

In individuals with normal vestibular function, the off-axis SVV tilts symmetrically during unilateral centrifugation. When the individual is offset to the right side of the axis of rotation, the SVV is tilted toward the left; and when the subject is offset to the left side of the axis of rotation, the SVV is tilted in a similar magnitude to the right. Patients with chronic unilateral vestibular loss exhibit an SVV asymmetry when measured during off-axis rotation. When the lesioned ear is centrifuged, the SVV does not shift because the utricle does not respond to the GIF. The SVV during unilateral centrifugation, therefore, may be a more sensitive test of otolith organ function than the static SVV test.

Faith W. Akin, PhD, CCC-A, is director of the Vestibular/Balance Laboratory at the Veterans Affairs Medical Center, Mountain Home, Tenn., and associate professor of communicative disorders at East Tennessee State University. Contact her at faith.akin@va.gov.  

Owen D. Murnane, PhD, CCC-A,  is director of the Electrophysiology Laboratory at the Veterans Affairs Medical Center, Mountain Home, Tenn., and associate professor of communicative disorders at East Tennessee State University. Contact him at owen.murnane@va.gov.

cite as: Akin, F. W.  & Murnane, O. D. (2009, February 10). Clinical Assessment of Otolith Function. The ASHA Leader.


GIF—gravito-inertial force; the sum of gravitational force and inertial force due to linear acceleration; the otolith organs are GIF sensors.

SCM—sternocleidomastoid muscle; VEMPs are recorded from surface electrodes placed on the SCM muscle.

SVV—subjective visual-vertical; SVV is a psychophysical measure of the angle between perceptual vertical and true (gravitational) vertical; SVV has been used as a test of otolith (utricular) function.

VEMP—vestibular-evoked myogenic potential; VEMPs are short-latency electromyograms used to test otolith (saccular) function.

The VEMP waveform is characterized by a positive peak (P1) at ~13 ms and a negative peak (N1) at ~23 ms post-stimulus onset. P1-N1 amplitudes are calculated from each ear, and inter-ear differences are expressed as an asymmetry ratio:

Asymmetry Ratio = 100 (AL – AS)/(AL + AS) in which AL equals the larger P1-N1 amplitude and AS equals the smaller P1-N1 amplitude. 



Akin, F.W., & Murnane, O. D. (2001). Vestibular-evoked myogenic potentials: preliminary report. Journal of the American Academy of Audiology, 12, 445–452.

Akin, F.W., Murnane, O. D., & Medley, T. M. (2003). The effects of click and tone burst stimulus parameters on the vestibular evoked myogenic potential (VEMP). Journal of the American Academy of Audiology, 14(9), 500–508.

Akin, F. W., Murnane, O. D., Panus, P. C., Caruthers, S. K., Wilkinson, A. E., & Proffit, T. M. (2004). The influence of voluntary tonic EMG level on the vestibular evoked myogenic potential. Journal of Rehabilitation Research & Development, 41(3B), 473–480.

Basta, D., Todt, I., Eisenschenk, A., & Ernst, A. (2005). Vestibular evoked myogenic potentials induced by intraoperative electrical stimulation of the human inferior vestibular nerve. Hearing Research, 204, 111–114.

Bohmer, A., & Rickenmann, J. (1995). The subjective visual vertical as a clinical parameter of vestibular function in peripheral vestibular diseases. Journal of Vestibular Research, 5, 35–45.

Colebatch, J. G. (2001). Vestibular evoked potentials. Current Opinion in Neurology, 14(1), 21–26.

Colebatch, J. G., & Halmagyi, G. M. (1992). Vestibular evoked potentials in human neck muscles before and after unilateral vestibular deafferentation. Neurology, 42, 1635–1636.

Curthoys, I. S., Kim, J., McPhedran, S. K., & Camp, A. J. (2006). Bone conducted vibration selectively activates irregular primary otolithic vestibular neurons in the guinea pig. Experimental Brain Research, 175(2), 256–267.

Furman, J. R., Schor, R. H., & Schumann, T. L. (1992). Off-vertical axis rotation: A test of the otolith-ocular reflex. Annals of Otology, Rhinology, and Laryngology, 101, 643–650.

Li, M. W., Houlden, D., & Tomlinson, R. D. (1999). Click evoked EMG responses in sternocleidomastoid muscles: Characteristics in normal subjects. Journal of Vestibular Research, 9, 327–334.

McCue, M. P., & Guinan, J. J. (1995). Spontaneous activity and frequency selectivity of acoustically responsive vestibular afferents in the cat. Journal of Neurophysiology, 74(4), 1563–1572.

Moffat, A. J. M., & Capranica, R. R. (1976). Auditory sensitivity of the saccule in the American toad (Bufo americanus). Journal of Comparative Physiology, 105, 1–8.

Popper, A. N., & Fay, R. R. (1973). Sound detection and processing by teleost fishes: A critical review. Journal of the Acoustical Society of America, 53(6), 1515–1529.

Rosengren, S. M. , Todd, N. P. M., & Colebatch, J. G. (2005). Vestibular-evoked extraocular potentials produced by stimulation with bone-conducted sound. Clinical Neurophysiology, 116, 1938–1948.

Sheykholeslami, K., Murofushi, T., Kermany, M. H., & Kaga, K. (2000). Bone-conducted evoked myogenic potentials from the sternocleidomastoid muscle. Acta Oto-Laryngologica, 120, 731–734.

Stockwell, C. W. (2000). Incidence of ENG abnormalities. Insights in Practice: Clinical Topics in Neurotology, May 2000.

Vanspauwen, R., Wuyts. F. L., & Van de Heyning, P. H. (2006). Improving vestibular evoked myogenic potential reliability by using a blood pressure manometer. The Laryngoscope, 116, 131–135.

Welgampola, M. S., & Colebatch, J. G. (2001b). Vestibulocollic reflexes: Normal values and the effect of age. Clinical Neurophysiology, 112, 1971–1979. 


Advertise With UsAdvertisement