BEDSIDE EXAMINATION OF THE VESTIBULAR SYSTEM

Introduction:

As noted earlier, the vestibular system works in conjunction with the visual and somatosensory

systems to achieve ocular and postural stability. To examine the vestibular system adequately,

one must isolate it from these other sensory systems. Historically this has been a difficult

task at the bedside, which is why most general textbooks only describe ways to assess the

“audio” component of the audio-vestibular system or eighth cranial nerve. The contribution

of the visual system can be removed with eye closure, but then the eye movements generated by the vestibular system cannot be observed. There is no simple way to eliminate somato sensation. Unlike the visual system, in which the optic nerve can be directly visualized and acuity accurately measured, the inner ear is located deep within the temporal bone, and the subjective

sensation to vestibular stimulation is ill defined. One can visualize the tympanic membrane

and some structures within the middle ear, but this generally provides no information about the status of the inner ear.

EXAMINATION OF THE EAR

The neurotologist should be familiar with the normal anatomy of the external canal and tympanic

Membrane, be capable of removing cerumen that interferes with visualization of the tympanic membrane, and be able to recognize certain common disorders on inspection (Fig. 1 ). Otoscopy is performed with the largest speculum that fits comfortably into the external canal: the pinna is gently pulled posterior and superior to straighten the canal. The tympanic membrane is normally

translucent; changes in color indicate middle ear disease (e.g., an amber color with middle ear effusions). Tympanosclerosis, the consequence of a resolved otitis media or trauma, appears as a semicircular crescent or horseshoe shaped white plaque within the tympanic membrane. It is rarely associated with hearing loss but is an important clue to past otitic infections. The pars

flaccida region, the area superior to the lateral process of the malleus, should be carefully inspected for evidence of a retraction pocket or attic cholesteatoma. The ossicles and the color of the underlying mucous membrane of the middle ear can often be assessed through a normal translucent tympanic membrane. Pneumatoscopy allows one to determine the mobility of the tympanic membrane. Lack of mobility may indicate an unsuspected perforation (usually under an anterior overhang), fl uid in the middle ear, or severe scarring of the tympanic membrane or

middle ear.

Figure 1. Appearance of the tympanic membrane in a normal subject (A) and in patients with a superior marginal perforation and cholesteatoma (B), a tympanic glomus body tumor (C), and a step deformity caused by a longitudinal temporal bone fracture (D).

Fistula Test

A fistula test is performed by transiently increasing and decreasing the pressure in the external canal with a pneumatoscope. A positive fistula sign (a transient burst of nystagmus and vertigo) occurs in patients with a perforated tympanic membrane and erosion of the bony labyrinth from conditions such as chronic infection, surgery, or trauma. The change in pressure is transmitted directly to the perilymph, compressing the membranous labyrinth and stimulating the semicircular canal cristae. The resulting nystagmus usually lasts from 10 to 20 sec. The direction of the nystagmus may be toward or away from the involved ear and is often the same for both positive and negative pressure changes. Lesser ocular and subjective responses may occur in patients with an intact tympanic membrane. Hennebert  first described this sign in patients with congenital syphilis, but it can occur in patients with a wide range of labyrinthine disorders. The response is a slow ocular deviation followed by a few beats of nystagmus even with sustained pressure. The abnormal eye movements are in the plane of the affected semicircular canal. With semicircular canal dehiscence syndrome the pressure changes are transmitted to the abnormal canal due to the presence of a “third window” in the bony labyrinth. In the case of endolymphatic hydrops (idiopathic or secondary to infection), fi brous adhesions between the medial surface of the stapedial footplate and the membranous labyrinth could result in displacement of endolymph when the footplate moves. This sign can occasionally be elicited during routine pneumatoscopy in normal subjects. In these cases it is usually present bilaterally. The mechanism by which pressure changes in the external auditory canal result in a pressure gradient across a normal vestibular receptor organs is unclear.

Fistulas of the labyrinth can also be tested for by inducing pressure changes by pressing and releasing the tragus (induces less pressure changes in the external canal than pneumatoscopy), Valsalva maneuver against closed glottis (increases intracranial pressure which can be transmitted to the inner ear by an intracranial fistula such as superior canal dehiscence), and Valsalva maneuver against pinched nostrils (increases pressure transmitted to the middle ear via the Eustachian tube).

v  TESTS OF VESTIBULOSPINAL REFLEXES

The labyrinths influence spinal cord motoneurons through the lateral vestibulospinal tract, the reticulospinal tract, and the descending medial longitudinal fasciculus (MLF). Labyrinthine stimulation of the spinal cord increases extensor tone and decreases flexor tone, resulting in a facilitation of the antigravity muscles. Both otolith and semicircular canal signals influence spinal cord anterior horn cells, but the former are more important in maintaining posture.

1.      Pastpointing

Pastpointing refers to a reactive deviation of the extremities caused by an imbalance in the

vestibular system (Fig. 2 ). The test is performed by having patients place their extended

index finger on that of the examiner’s, close their eyes, raise the extended arm and index finger to a vertical position, and attempt to return the index finger to the examiner’s. Consistent deviation to one side is pastpointing.

Figure 2. Bedside tests of vestibulospinal function.

Pastpointing tests represent one of the earliest attempts to clinically assess vestibular function. In 1910, Barany 5 published a review of pointing deviation and emphasized the importance of having patients sit with their eyes closed to avoid confusion with other orienting information. Barany showed that caloric stimulation consistently induced pastpointing in the direction of the slow component of induced nystagmus. Cold caloric irrigation (inhibiting the horizontal ampullary nerves’ spontaneous firing rate) resulted in pastpointing toward the irrigated ear, and warm caloric irrigation induced the opposite effect. As expected, patients with acute unilateral loss of vestibular function past-pointed toward the damaged side. Barany and numerous others emphasized that repeated testing shows a large variability, occasionally with drift in the wrong direction. Subsequent investigators tried to improve test accuracy by eliminating tactile feedback and using small finger lamps that could be photographed, but the large variability among normal subjects and patients remained. It is apparent that results from a single pointing test can be misleading and should not be considered clinically relevant when in isolation. Extra labyrinthine influences should be eliminated as much as possible by having the patient seated with eyes covered and arms and index fingers extended throughout the test. The standard finger-to-nose test will not identify pastpointing, inasmuch as joint and muscle proprioceptive signals permit accurate localization even when vestibular tone is asymmetric. Although patients with acute peripheral vestibular damage usually past-point toward the side of loss, compensation rapidly corrects the pastpointing and can even produce a drift to the other side.

2.      Static Posture

Patients with damage to the vestibular system often suffer instability of the trunk and lower limbs so that they sway back and forth or even fall to one side. In 1846, Romberg 6 noted that

patients with proprioceptive loss from tabes dorsalis were unable to stand with feet together and eyes closed. Barany 7 first emphasized the importance of vestibular influences in maintaining the Romberg position (Fig. 2 ). As with pastpointing, he noted that patients with acute unilateral labyrinthine lesions swayed and fell toward the diseased side, that is, in the direction of the slow component of nystagmus. However, like the pastpointing test, the Romberg test was found to be rather insensitive for detecting chronic unilateral vestibular impairment, and sometimes the patient would fall toward the intact ear. The so-called sharpened Romberg test is a more sensitive indicator of vestibular impairment. For this test the patient stands with feet aligned in the tandem heel-to-toe position with eyes closed and arms folded against the chest. Most normal subjects under the age of 70 can stand in this position for 30 sec: older normal subjects and patients

with unilateral or bilateral vestibular impairment usually cannot sustain the position. Abnormalities on this test, however, are not specific to the vestibular system. Although lower mammals consistently develop ipsilateral hypotonia of extensor muscles after labyrinthectomy, one rarely finds this in human patients. Occasionally, slight asymmetry in posture is found with the ipsilateral upper extremity slightly flexed and abducted compared to the contralateral upper extremity. The clinically elicited deep tendon reflexes are also unaffected by vestibular lesions.

Apparently, other supraspinal influences on the anterior horn cells rapidly compensate for the loss of tonic vestibular signals.

3.      Walking Tests

Unterberger was the first to systematically study the tendency of vestibular stimulation or

unilateral vestibular lesions to induce blindfolded subjects to turn in the earth’s vertical axis when walking. The direction of turning coincided with the direction of pastpointing and falling (in the direction of the slow component of nystagmus). Fukuda 9 obtained similar results by having subjects take 50 to 100 steps on the same spot and recording the angle of rotation as well as forward and backward movements. Both of these tests were performed with arms extended parallel and horizontal in front of the subject, so upper extremity deviation (pastpointing) may have added to the tendency to rotate in a given direction. Tandem gait tests (Fig. 2 ) are widely used as part of the routine neurologic examination, and most clinicians recognize normal and abnormal performances. When performed with eyes open, tandem walking is primarily a test of cerebellar function, because vision compensates for chronic vestibular and proprioceptive deficits. Acute vestibular lesions, however, typically impair tandem walking, even with the eyes open. Recently, there has been an emphasis on timed walking tests to provide a semiquantitative

measure of balance and risk of falling in older patients. These tests are not specific for vestibular function but rather provide an overall measure of gait and balance. Gait speed (over a set path), timed tandem stance and walking, and maximum step length (ability to maximally step out and return to the initial position) have all been shown to predict the likelihood of falls in older people.

v  TESTS OF VESTIBULO-OCULAR REFLEXES

Doll’s Eye Test (Oculocephalic Response)

The doll’s eye test involves slowly moving the head back and forth in the horizontal plan to

induce reflex eye movements. In an alert human these eye movements result from combined

visual and vestibular stimulation. Therefore, a patient with complete loss of vestibular function will have compensatory eye movements on the test if the visuomotor system is intact. On the other hand, an alert human can also use the visuomotor system (i.e., the smooth pursuit system) to overcome or suppress the vestibular eye movement in this test that uses slow passive movements of the head. In a comatose patient, however, the doll’s eye test is a useful bedside test of the vestibuloocular reflex since the pursuit system is not functioning. Conjugate compensatory eye movements indicate normally functioning vestibulo-ocular pathways. This test is a standard component of the coma exam and also brain death exam because the reflex eye movements indicate intact function not only of the inner ear but also the brain stem. Absence of reflex eye movements in a comatose patient is usually an ominous sign, indicating massive brain stem damage if acute drug intoxication or metabolic disorders can be ruled out.

The doll’s eye test can also provide highly localizing information. For example, if the test results in disconjugate eye movements, then one could localize the lesion to MLF, the oculomotor neurons, or the ocular muscles (depending on the abnormal pattern).

Head-Thrust Test

The head-thrust test is a simple way to identify a complete unilateral or bilateral loss of vestibular function at the bedside. The discovery of this test was a major breakthrough in the

ability to examine the vestibular system at the bedside. Prior to the head-thrust test, the bedside

assessment was mostly limited to searching for surrogate signs of an acute vestibular imbalance (i.e., nystagmus). Tests such as pastpointing, the Unterberger test, and the doll’s eye test could be used, but these had major limitations as noted previously. The head-thrust test is used to directly assess for vestibular de-afferentation and thus is analogous to the test for an afferent pupillary defect (i.e., the swinging light test). Though the afferent pupillary defect was first reported around 1900, the head-thrust test was not reported until 1988.  Despite the consensus opinion

among neurotologists regarding the substantial value of the head-thrust test, it remains a test

that is underappreciated and underutilized in the general medical community. The head-thrust test is performed by grasping the patient’s head and applying brief, small amplitude,

high-acceleration head thrusts, first to one side and then to the other (Fig. 3 ). Before the movement, the patient is instructed to fixate on the examiner’s nose. During and after the quick

movements, the examiner closely watches the patient’s eye position looking specifically for

“catch-up” saccades, which are the sign of an inappropriate compensatory vestibular system

response. If the vestibular system is intact on both sides, then movement to either side triggers

the eye movements (the eye movements are triggered by a reflex, the vestibulo-ocular

reflex) that keep the patient’s eyes fixated on the examiner’s nose. On the other hand, if a vestibular lesion is present, then the movement in the direction of the lesion results in the eyes

moving with the patient’s head (because the reflex is not working to move the eyes in the

opposite direction). Thus, if a patient has a right vestibular lesion, then after a head thrust to the

right, the patient’s eyes will move with the head (stay in midline position), and then the patient

will have to make a voluntary saccade (“catchup” or “corrective” saccade) back to the examiner’s nose. The passive head movement is fast enough (high enough acceleration) that the smooth pursuit system cannot be used to keep the patient’s eyes fixated on the examiner’s nose

(recall how the smooth pursuit system can cover up for vestibular impairment with the low velocity and low-frequency movements of the doll’s eye test). In addition, the patient also cannot

initiate a voluntary saccade fast enough to keep the eye’s on the nose and for this reason a

“catch-up” saccade is required to move the eye back to the nose. Of course, if a patient can correctly predict the direction of the head movement and then time the initiation of a saccade

accurately, then a voluntary saccade can come very close to covering up for a vestibular lesion.

In the same way that the afferent pupillary defect is mostly a test of the optic nerve function,

the head-thrust test is mostly a test of the vestibular nerve function. Intact end-organ

(e.g., labyrinthine) function is of course required to generate the vestibulo-ocular reflex

but a disorder of the end-organ may not result in a positive test unless it is severe or complete.

Similarly, retinal lesions, unless severe, do not typically result in an afferent pupillary defect.

For this reason, the head-thrust test is typically not positive in patients with Meniere’s disease

unless a procedural lesion has been performed. In addition, the accuracy of the headthrust

test is also influenced by the severity of the nerve lesion. For example, the head-thrust

test may not be positive if the vestibular nerve lesion is only mild to moderate in severity (i.e.,

resulting in approximately less than a 50 % paresis) but is nearly always positive with lesions that are moderate to severe. It is not completely fair to gauge the value of the qualitative positive versus negative results of the head-thrust test to the quantitative results of the caloric test. The reason is because the head thrust is a test of the vestibular system at high acceleration, whereas the caloric test is a test of the vestibular system at a very low acceleration. Furthermore, the caloric test, though generally regarded as the gold standard test for a unilateral vestibular lesion, is known to have concerning levels of false-positive results based on the limitations of the test and thus is not an optimal gold standard test.  The head-thrust test is nearly always positive in patients with acute vestibular neuritis because the lesion is of the vestibular nerve and is typically at least a moderate to severe lesion. For this reason, the head thrust is particularly valuable in the presentation of the acute vestibular syndrome because a positive head-thrust test is a very strong indicator of a lesion of the vestibular nerve (and thus the most common cause, vestibular neuritis). A negative head-thrust test in the acute vestibular syndrome suggests the vestibular nerve is intact and thus the likelihood of a brainstem or cerebellar lesion increases.  The head-thrust test is also particularly helpful for identifying a bilateral vestibulopathy. A bilateral vestibulopathy can be difficult to recognize based on the symptom report and general examination. As a result, many patients likely go undiagnosed. However, a positive bilateral head-thrust test can clinch the diagnosis at the bedside.

Figure 3. The head thrust test. The head thrust test is a test of vestibular function that can be easily done during the bedside examination. This maneuver tests the vestibulo-ocular refl ex (VOR). The patient sits in front of the examiner and the examiner holds the patient’s head steady in the midline. The patient is instructed to maintain gaze on the nose of the examiner. The examiner then quickly turns the patients head about 10-15 degrees to one side and observes the ability of the patient to keep the eyes locked on the examiner’s nose. If the patient’s eyes stay locked on the examiner’s nose (i.e., no

corrective saccade) (picture a ), then the peripheral vestibular system is assumed to be intact . If, however, the patient’s eyes move with the head (picture b ) and then the patient makes a voluntary eye movement back to the examiner’s nose (i.e., corrective saccade), then this indicates a lesion of the peripheral vestibular system and not the central nervous system . Thus, when a patient presents with the acute vestibular syndrome, the test result shown in picture A would suggest a CNS lesion (because the VOR is intact), whereas the test result in picture B would suggest a peripheral vestibular lesion(because the VOR is not intact).

Dynamic Visual Acuity

The dynamic visual acuity test is performed by having the patient shake the head rapidly back

and forth in the horizontal plane at approximately 2 Hz while reading a Snellen visual acuity chart at the standard distance. Inasmuch as the smooth pursuit system functions best below 1 Hz and almost not at all at 2 Hz, this is primarily a test of the horizontal vestibuloocular reflex. A drop in acuity of more than one line on the Snellen chart suggests an abnormal vestibulo-ocular reflex. The test is most useful for identifying bilateral vestibular loss but can also be abnormal with unilateral vestibular loss and with cerebellar lesions.

Cold Caloric Test

Because of its ready availability, ice water (approximately 0 ° C) is commonly used for bedside caloric testing. 27 Tap water can be brought to 4 ° C in about 10 minutes after adding ice cubes. To bring the horizontal canal into the vertical plane, the patient lies in the supine position with the head tilted 30 degrees forward or in the sitting position with the head tilted 60 degrees backward . Infusion of ice water induces a burst of nystagmus with slow phase toward the side of infusion and the fast phase in the opposite direction, usually lasting from 1 to 3 min. The volume of ice water recommended for this test varies from 0.5 to 2 ml. Regardless of the volume used, however, it is critical that the stimulus reach the eardrum (i.e., not be injected into the canal wall or in a canal blocked by cerumen). Direct visualization of the eardrum is mandatory. We suggest using 1 or 2 ml of ice water infused directly against the tympanic membrane through a small rubber hose. The ear being infused is turned upward for approximately 30 sec after the infusion to be certain that the water stays against the drum. In an alert subject, a burst of nystagmus will develop within 30 sec to 1 min after infusion and last from 1 to 3 min. In a comatose patient, only a slow tonic deviation toward the side of stimulation is observed because the brain does not generate the fast phases. In normal subjects, duration and speed of induced nystagmus varies greatly, but > 20 % asymmetry in nystagmus duration suggests the possibility of a lesion on the side of the decreased response, though this test is less valid than standard bithermal caloric testing with nystagmography.

Rotational Testing

Qualitative rotational testing of the horizontal vestibulo-ocular reflex can be performed at the

bedside by using a swivel chair. Barany  introduced a rotatory test in which the chair on which the patient was seated was manually rotated 10 times in 20 sec and then suddenly stopped with

the patient facing the observer. The duration of postrotatory nystagmus in each direction was

then measured. In normal subjects an average of 22 sec was required for cessation of postrotatory

nystagmus, but intersubject variability was large. Much of this variability could be traced to the difficulty in manually maintaining constant velocity and then a uniform sudden deceleration. Furthermore, the vestibular response to the initial acceleration was often not completed before deceleration began, resulting in interaction between the two responses. As with the ice water caloric testing, this type of qualitative testing can provide only gross information about the presence and symmetry of vestibular function, and thus it is not felt to be as valuable as the head-thrust test. One aspect of rotational testing that is useful at the bedside is the fixation-suppression test. With this test the subject extends the arm rigidly and attempts to fixate on the extended

thumb while the entire body is rotated back and forth en bloc. Normal subjects can completely

suppress their vestibulo-ocular reflex, keeping their eyes fixed in the center of the orbits. Abnormal fixation-suppression (nystagmus) indicates impairment of the smooth pursuit systems and thus is an indicator of a central lesion, often involving the cerebellum.

TESTS FOR PATHOLOGIC NYSTAGMUS

Nystagmus can be defined as a non voluntary rhythmic oscillation of the eyes. It usually has

clearly defined fast and slow components alternating in opposite directions. By convention,

the direction of the fast component defines the direction of nystagmus (e.g., left-beating nystagmus indicates the fast phase is to the left). Physiologic nystagmus refers to nystagmus that

occurs in normal subjects, whereas pathologic nystagmus implies an underlying abnormality

(Table 1 ). Spontaneous nystagmus refers to nystagmus that occurs with the eyes in the primary

position, without external stimulation such as movement of the head or surroundings. Gaze-evoked nystagmus is induced by changes in gaze position. Nystagmus that is not present in the sitting position but is present in some other head and body position is called positional nystagmus . This definition excludes nystagmus present in the sitting position that is modified by a change in position.

Table 1:  Types of Nystagmus

Methods of Examination:

The clinical examination for pathologic nystagmus should include a systematic study of changes in

( 1 ) fixation,

( 2 ) eye position, and

( 3 ) head position.

Omission of any of these three maneuvers may lead to overlooking the presence of nystagmus or misinterpreting its type. Sometimes pathologic nystagmus can be triggered by vibration, head shaking, or hyperventilation. Spontaneous nystagmus may be present with fixation, or it may occur only when fixation is inhibited . There are several simple methods for achieving the latter at the bedside. Frenzel glasses consist of + 30 lenses mounted in a frame that contains a light source on the inside so that the patient’s eyes are easily visualized (Fig. 4 ). The light can be powered by a battery, making the entire system portable. Frenzel glasses should be used

only in a darkened room, because the patient can fixate (at least partially) through the lenses

in a lighted room. An ophthalmoscope can also be used to block fixation and bring out a

spontaneous nystagmus. While the fundus of one eye is being visualized the patient is asked

to lightly cover the other eye with one hand. Nystagmus appears as a slow drift of the retina

in one direction interrupted by flicking movements in the opposite direction (the direction

of the nystagmus is reversed, inasmuch as one is visualizing the back pole of the eye).

Alternatively, one can simply shine a penlight in one eye while intermittently occluding the

other eye.  In some cases, simply holding a blank sheet of paper in front of the patient’s

vision and asking the patient to stare through it is enough to bring out the spontaneous nystagmus. When this is done, the examiner has to look at the patient’s eyes from the side.

Figure 4. Frenzel glasses.

Occasionally, nystagmus can be seen even through closed lids. This can be misleading,

however, because lid-twitch movements often mimic nystagmus. The effect of change in eye position is evaluated by having the patient fi xate on a target 30 degrees to the right, left, up, and down. Because horizontal eye deviation beyond 40 degrees may result in a low-amplitude, high-frequency torsional nystagmus in normal subjects (socalled end-point or end-gaze nystagmus),

extreme eye positions should be avoided. Each eye position is held for at least 20 sec. First degree nystagmus refers to nystagmus that is present only on gaze in the direction of the fast

component. Second-degree nystagmus is present in the midposition (primary position) and on gaze in the direction of the fast component, and third-degree nystagmus is present even on

gaze away from the fast component. These terms are not applicable to all varieties of nystagmus

and, therefore, can lead to confusion. A simple description can be rapidly summarized with a box diagram as illustrated in Figure 5 . The size, shape, and direction of the arrows provide information about the amplitude and direction of the fast component of nystagmus in each eye position. Routinely, two types of positional testing are used: slow and rapid. With the first, the patient slowly moves into the supine, right lateral, and left lateral positions. Positional nystagmus

induced by slow positioning is persistent, low in frequency, and often present only when fixation

is inhibited. Paroxysmal positional nystagmus, however, is induced by a rapid change

from the erect sitting to the supine headhanging left (left Dix-Hallpike test), center, or right position (right Dix-Hallpike test) or from the supine to head-right or head-left position

(supine positional testing). It is initially high frequency but rapidly dissipates within 30 sec to 1 min.

Figure 5. Method for describing the effect of eye position on nystagmus amplitude and direction. Arrows indicate direction of nystagmus (direction of fast component) in each eye position.

For vibration-induced nystagmus, a handheld vibrator (approximately 100 Hz) is placed

on the mastoid and suboccipital bones on each side for 10 to 15 seconds. The test for headshaking nystagmus is performed by having the patient shake the head back and forth in the

horizontal and vertical planes for approximately 10 cycles. For hyperventilation-induced nystagmus, the patient breathes rapidly in and out for about a minute and a half.

TYPES OF PATHOLOGIC NYSTAGMUS

A.    Spontaneous Nystagmus

mechanism : Spontaneous nystagmus results from an imbalance of tonic signals arriving at the oculomotor neurons. Because the vestibular system is the main source of oculomotor tonus, it is the driving force of most types of spontaneous nystagmus (tonic signals arising in the pursuit and

optokinetic systems may also play a role, particularly with congenital nystagmus). A vestibular

imbalance results in a constant drift of the eyes in one direction (slow phase) interrupted

by fast components in the opposite direction. If the imbalance results from a peripheral vestibular lesion, patients can use their pursuit system to cancel it. If it results from a central vestibular lesion, their pursuit system usually cannot suppress it (because central vestibular and pursuit pathways are highly integrated; see “Visual–Vestibular Interaction” ). The features that separate

peripheral from central varieties of spontaneous nystagmus are summarized in Table 2 .

Table 2:  Differentiation between Spontaneous Nystagmus of Peripheral and Central Origin

        i.            PERIPHERAL SPONTANEOUS NYSTAGMUS

Lesions of the peripheral vestibular system (labyrinth or eighth nerve) typically interrupt

tonic afferent signals originating from all of the receptors of one labyrinth so that the

resulting nystagmus has combined torsional, horizontal, and vertical components. The horizontal

component dominates, because the tonic activity from the intact vertical canals and otoliths partially cancels out. Gaze in the direction of the fast component increases the frequency and amplitude, whereas gaze in the opposite direction decreases the frequency and amplitude (Alexander’s law)  30 Peripheral spontaneous nystagmus is “unidirectional” meaning that the primary direction does not change. Thus, if the primary position nystagmus is right-beating, then it will not change to left beating nystagmus even on left gaze. The slow phase is linear, resulting in a saw-toothed waveform. As noted previously, peripheral spontaneous nystagmus is strongly inhibited by fixation. Unless the patient is seen within a few days of the acute episode, spontaneous nystagmus will not be present when fixation is permitted.

      ii.            CENTRAL SPONTANEOUS NYSTAGMUS

Central spontaneous nystagmus is usually prominent with and without fi xation. It may be

purely vertical, horizontal, or torsional, or have some combination of torsional and linear

components . As with peripheral spontaneous nystagmus, gaze in the direction of the fast component usually increases nystagmus frequency and amplitude, but unlike peripheral spontaneous nystagmus, gaze away from the direction of the fast component will often change the direction of the nystagmus (i.e., direction changing nystagmus). There is typically a null

region several degrees off center in the direction opposite that of the fast component where

nystagmus is minimal or absent. Gaze beyond this null region results in reversal of nystagmus

direction. The slow phase of central spontaneous nystagmus may be linear, exponentially

increasing, or exponentially decreasing.  With spontaneous downbeat nystagmus the vertical amplitude increases with horizontal gaze deviation.  Downward gaze also increases

the amplitude in about two-thirds of cases, but in the other one-third it decreases it. Upward

gaze may reverse the direction to upbeat. Downbeat nystagmus has been produced in monkeys after lesions of the uvula and flocculonodular lobes of the cerebellum. In the human it is localizing to the cervicomedullary junction (which includes the midline cerebellar regions). Common causes of downbeat nystagmus include cerebellar atrophy, vertebrobasilar ischemia, multiple sclerosis, and Arnold-Chiari malformation. Central spontaneous nystagmus has generally been attributed to an imbalance in either central vestibulo-ocular or smooth pursuit

pathways. Horizontal and vertical pathways are usually separate so that lesions can commonly

lead to pure horizontal or pure vertical nystagmus. Often central spontaneous nystagmus

is altered by position change, suggesting that peripheral otolith input can alter the central

imbalance. Marti and colleagues suggested that downbeat nystagmus results from damage

to the inhibitory vertical gaze-velocity sensitive Purkinje cells in the cerebellar flocculus.

These neurons have spontaneous activity and most exhibit downward on-directions. Loss of

these vertical flocculus Purkinje cells would lead to disinhibition of their brainstem target neurons and spontaneous upward drift (i.e., downbeat nystagmus).

B.     CONGENITAL NYSTAGMUS

mechanism : Congenital spontaneous nystagmus is almost always highly dependent on fi xation, disappearing

or decreasing with loss of fixation. In some instances a slow nystagmus in the reverse direction

occurs when fixation is inhibited. One common variety, so-called latent congenital nystagmus, occurs only when either eye is covered, permitting monocular fixation. The resulting nystagmus beats toward the fixating eye. Latent congenital nystagmus is usually associated with other congenital ocular defects such as concomitant squint and alternating hyperphoria.  Several characteristic clinical features help distinguish congenital from acquired spontaneous nystagmus. It is usually purely horizontal and may diminish or disappear with convergence. The waveform may be pendular to sawtooth shaped, with many variations in between.  Different waveforms occur in the same patient in different eye positions. Gaze in the direction of the fast component converts a pendular nystagmus to a jerk nystagmus; often there is a null region where the nystagmus is minimal. Several different waveforms may be seen in members of the same family with congenital nystagmus. The frequency of congenital nystagmus is usually > 2 beats/sec and at times reaches 5 to 6 beats/sec. Nystagmus of this high frequency is unusual other than on a congenital basis. Of course, most patients are aware that the nystagmus has been present since infancy. The pathophysiologic mechanism of congenital nystagmus is only partially understood.

Convincing evidence exists that the slow component causes the target to slip from the fovea,

and the fast component brings the target back to the fovea. The slow component is not the result of, but the cause of, decreased vision. Maneuvers designed to decrease the target slippage (fitting glasses with prisms and extraocular muscle surgery) improve visual acuity. Patients with congenital nystagmus can make normal-velocity saccades, indicating that the extraocular muscles and orbital mechanics are normal. The vestibular system also appears to be normal in most of these patients. Abnormalities in smooth pursuit and optokinetic slow phases are uniformly present, but it is difficult to know whether these abnormalities are due to a superimposition of the spontaneous nystagmus on attempted tracking eye

movements or to an underlying abnormality. Recently mutations have been identified in two

genes associated with X-linked congenital nystagmus: FRMD7 causing X-linked idiopathic congenital nystagmus and GPR143 causing X-linked ocular albinism and congenital nystagmus.

C.    Gaze-evoked nystagmus

Mechanism : Patients with gaze-evoked nystagmus are unable to maintain stable conjugate eye deviation away

from the primary position. The eyes drift back toward the center with an exponentially decreasing wave form; corrective saccades (fast components) constantly reset the desired gaze position. Gaze-evoked nystagmus is therefore always in the direction of gaze. The site of

abnormality can be anywhere from the neuromuscular junction to the multiple brain centers

controlling conjugate gaze (Table 3 ). Dysfunction of the oculomotor integrator may be a common mechanism for several types of gaze-evoked nystagmus.

Table 3 Causes of Gaze-Evoked Nystagmus

a)      SYMMETRIC

Symmetric gaze-evoked nystagmus (equal amplitude to the left and right) is most commonly produced by ingestion of drugs such as phenobarbital, phenytoin, alcohol, and diazepam. With these agents, high-frequency, smallamplitude nystagmus (<2 degrees) is found in all directions of gaze. A rough correlation exists between nystagmus amplitude and blood drug level. The nystagmus initially appears at extreme horizontal gaze positions and moves toward the midposition with higher drug levels. In addition to its association with drug ingestion, symmetric gaze-evoked nystagmus commonly occurs in patients with myasthenia gravis, multiple sclerosis, and cerebellar atrophy.

b)     ASYMMETRIC

Asymmetric horizontal gaze-evoked nystagmus always indicates a structural brain lesion. When it is caused by a focal lesion of the brain stem or cerebellum, the larger amplitude nystagmus is

usually directed toward the side of the lesion. Large cerebellopontine angle tumors commonly

produce asymmetric gaze-evoked nystagmus from compression of the brain stem and cerebellum (Bruns’ nystagmus). Some patients with large acoustic neuromas develop a combination of asymmetric gaze-evoked nystagmus from brainstem compression and peripheral spontaneous nystagmus from eighth nerve damage. Asymmetric gaze-evoked nystagmus may be present during the recovery from gaze paralysis (either cortical or subcortical in origin), in which case it is large in amplitude and low in frequency and present only in one direction of gaze (the direction of the previous gaze paralysis).

c)      REBOUND

Rebound nystagmus is a type of gaze-evoked nystagmus that either disappears or reverses

direction as the lateral gaze position is held. When the eyes return to the primary position,

another burst of nystagmus occurs in the direction of the return saccade. Thus, the patient

may have a transient primary position nystagmus in either direction. Rebound nystagmus

occurs in patients with cerebellar atrophy and focal structural lesions of the cerebellum; it is

the only variety of nystagmus thought to be specific for cerebellar involvement.\

d)     DISSOCIATED

Dissociated, or disconjugate, gaze-evoked nystagmus commonly results from lesions of the

medial longitudinal fasciculus (MLF), so-called internuclear ophthalmoplegia. With early MLF

lesions the eyes appear to move conjugately, but the abducting eye on the side opposite the

MLF lesion develops a regular small-amplitude, high-frequency nystagmus in the direction of

gaze. With more extensive MLF lesions, the adducting eye lags behind and develops a lowamplitude nystagmus while the abducting eye overshoots the target and develops largeamplitude nystagmus that has a characteristic peaked waveform.  A MLF nystagmus can be bilateral or unilateral, depending on the extent of MLF involvement. Bilateral MLF nystagmus is most commonly seen with demyelinating disease, whereas unilateral MLF nystagmus most often accompanies vascular disease of the brain stem. Patients with myasthenia gravis develop dissociated gaze-evoked nystagmus similar to MLF nystagmus (pseudo-MLF nystagmus) because of unequal impairment of neuromuscular transmission in adducting and abducting muscles. Unlike MLF nystagmus, the dissociated nystagmus with myasthenia progressively increases in amplitude as the gaze position is maintained.

D) Positional Nystagmus

Mechanism: Beginning with Barany, positional nystagmus was attributed to lesions of the otoliths and their connections in the vestibular nuclei and cerebellum, as these are the receptors that are sensitive to changes in the direction of gravity.  Subsequently, other mechanisms for the

production of positional nystagmus have been identified, forcing reexamination of these traditional concepts. If a semicircular canal cupula is altered so that its specific gravity no longer

equals that of the surrounding endolymph or if debris inappropriately enters a semicircular

canal, the canal becomes sensitive to changes in the direction of gravity and can produce

positional nystagmus.  Traditional classifications of positional nystagmus are often confusing and can be difficult to apply in clinical practice. Some classifications have been based on clinical observations obtained while the patient is fixating, whereas others have been based on electronystagmography (ENG) recordings with eyes closed or with eyes open in darkness. Some investigators use slow positioning maneuvers, but others employ only rapid positioning. These different methods make it difficult to compare classifications. Nylen  initially described three types of positional nystagmus based on visual inspection of nystagmus direction and regularity.

Type I, direction changing, and type II, direction fixed, remained constant as long as the

position was maintained. Type III was less clearly defined, comprising all paroxysmal varieties of positional nystagmus and some persistent varieties that did not fit into types I and II. Numerous modifi cations of Nylen’s original classification have subsequently been proposed,

and the definition of each type has changed. Most investigators do agree that two broad categories of positional nystagmus can be identified: paroxysmal and persistent.

1)      PAROXYSMAL POSITIONAL NYSTAGMUS (POSITIONING NYSTAGMUS)

The most common type of paroxysmal positional nystagmus is induced by a rapid change

from erect sitting to the supine head-hanging left or right position (Dix-Hallpike test) (Fig. 6 ).

Figure 6. Method for inducing paroxysmal positional nystagmus (Dix-Hallpike maneuver). Patient is taken rapidly from sitting to head-hanging right (a) or head-hanging left

(b) position.

It is initially high in frequency but dissipates rapidly. There is a 3- to 10-sec latency before onset and the nystagmus rarely lasts longer than 30 sec. The nystagmus has combined torsional (fast component toward the undermost ear) and vertical (fast component toward the forehead) components. Although infrequent bilateral cases do occur, the nystagmus is usually prominent only in one head-hanging position, and a burst of nystagmus occurs in the reverse direction when the patient moves back to the sitting position. Another key feature is that the patient experiences severe vertigo with the initial positioning, but with repeated positioning, vertigo and nystagmus rapidly disappear (fatigability). This type of paroxysmal positional nystagmus is specific for the

posterior canal variant of canalithiasis. Horizontal canal variants also exist but are less common. They are induced by turning the patient’s head to the side while the patient lies supine.

Paroxysmal positional nystagmus can also result from brainstem and cerebellar lesions.

The central type does not decrease in amplitude or duration with repeated positioning,

does not have a clear latency, and usually lasts longer than 30 sec. The direction is unpredictable and may be different in each position. It is often purely vertical with fast phase directed downward (i.e., toward the cheeks). The presence or absence of associated vertigo is not a reliable differential feature. Central paroxysmal positional nystagmus can be the initial presenting sign of a posterior fossa tumor such as a medulloblastoma or cerebellar glioma. It is therefore critical to distinguish it from the benign peripheral variety (Table 4 ).

Table 4 Differentiation between Peripheral and Central Paroxysmal Positional Nystagmus

2)      PERSISTENT POSITIONAL NYSTAGMUS

This type of positional nystagmus remains as long as the position is held, although it may

fluctuate in frequency and amplitude. It may be in the same direction in all positions or

change directions in different positions. Direction-changing and direction-fixed static

positional nystagmus are most commonly associated with peripheral vestibular disorders, although both occur with central lesions. One variety of persistent direction-changing positional nystagmus (apogeotropic) is thought to result from otolithic debris attached to the cupula or lodged in the ampulla of the horizontal semicircular canal . As with spontaneous nystagmus, lack of suppression with fixation and signs of associated brainstem dysfunction suggest a central lesion.

E) Head-Shaking Nystagmus

Patients with a compensated vestibular imbalance due to either peripheral or central lesions may develop a transient nystagmus after vigorous head shaking. With unilateral peripheral

vestibular lesions, the abnormal side is in the direction of the slow phase. With central vestibular

lesions, the direction of nystagmus is nonlocalizing. Sometimes vertical nystagmus is

induced by horizontal head shaking. The results of vertical head shaking are more difficult to interpret because some normal subjects will have transient vertical nystagmus after vertical head shaking.

F) Hyperventilation-Induced Nystagmus

Hyperventilation can induce a near-faint dizziness in anyone, particularly in anxious patients,

but hyperventilation-induced nystagmus is less common. Patients with compressive lesions of the vestibular nerve, such as with an acoustic neuroma or cholesteatoma, or with

demyelination of the vestibular nerve root entry zone, such as with multiple sclerosis, may develop nystagmus after hyperventilation. Presumably metabolic changes associated with hyperventilation trigger the partially damaged nerve to fire inappropriately. Hyperventilation-

induced nystagmus has also rarely been associated with labyrinthitis or perilymph fistula.

OTHER OCULAR OSCILLATIONS

·         Dissociated Spontaneous Nystagmus

 Several different lesions of the posterior fossa can result in a spontaneous nystagmus with torsional, horizontal, and vertical components varying in each eye. The nystagmus is usually

synchronized, however, in that the fast component occurs at exactly the same time in both

eyes. Tumors, vascular disease, and demyelinating disease of the brain stem produce this

form of dissociated nystagmus. Frequently the eye on the side of the lesion shows the largest

amplitude oscillation. Monocular nystagmus results from similar posterior fossa lesions;

this unusual form of dissociated nystagmus also has been reported with such varied entities as

congenital syphilis, meningitis, optic nerve glioma, cerebral trauma, unilateral amblyopia,

and high refractive error.As expected, these patients are typically bothered by severe

oscillopsia. Seesaw nystagmus is an unusual type of dissociated nystagmus in which one eye rhythmically rises and intorts and the other eye falls and extorts. It may be congenital but most

often is produced by acquired lesions near the optic chiasm, particularly those producing a

bitemporal fi eld defect and decreased central visual acuity. Lesions associated with seesaw

nystagmus include craniopharyngiomas, syringobulbia, brainstem infarction, and diffuse

choroiditis; compression of the midbrain tegmentum may be the common denominator.

·         Convergence Retraction Nystagmus

This dramatic ocular motor disorder results from lesions involving the diencephalic midbrain

junction. When the patient attempts to make voluntary upward saccades or when involuntary

upward saccades (fast components) are induced by an optokinetic or vestibular stimulus, the

patient develops co-contraction of all extraocular muscles and the eyes rhythmically retract

and converge. In other cases convergence nystagmus occurs without retraction, apparently because of asynchronous adducting saccades. Convergence retraction nystagmus is usually associated with other signs of midbrain dysfunction (impaired upward gaze, pupillary abnormalities, accommodative spasm, retraction of the lids, and skew deviation), constituting

the dorsal midbrain syndrome. This syndrome is most frequently produced by dysgerminomas

of the pineal region but is also associated with other tumors and vascular lesions involving the tectal or pretectal area.

·         Ocular Bobbing

Ocular bobbing consists of abrupt, nonrhythmic, conjugate, downward jerks of the eyes,

followed by slow return to midposition. The abnormal movements are classically seen in comatose patients with intrinsic pontine lesions that also produce absent reflex horizontal eye movements, but they have also been reported with posterior fossa lesions that compress the pons and with metabolic encephalopathy. Inverse ocular bobbing or ocular dipping refers to a slow downward movement of the eyes followed by a rapid return to midposition. Reverse bobbing consists of a rapid deviation of the eyes upward followed by a slow return to the primary position. These latter phenomena may be variations of ocular bobbing because all can be seen in the same subject at different times. As with typical ocular bobbing, they are usually seen with metabolic disorders or structural lesions of the pons.

·         OCULAR TILT REACTION

If a subject is tilted in the frontal plane (about the nasal occipital axis), the head reflexively tilts and the eyes counter-roll and skew toward the opposite side. The functional role of this reflex in visual stabilization during natural body movements is minimal, however, as the magnitude of the compensatory head tilt and ocular counter-rolling is only about 10 % of angular displacement of the head. The ocular tilt reaction is principally a labyrinthine reflex; it is independent of the position of the head, relative to the body (indicating that neck position is not important). The ocular tilt reaction can be elicited by electrical stimulation of the rostral midbrain tegmentum in the region of the interstitial nucleus of Cajal. Clinically, the ocular tilt reaction has been seen in patients with peripheral labyrinthine lesions (a complication of stapedectomy), lesions of the lateral medulla (particularly Wallenberg’s syndrome), and with lesions of the rostral midbrain.

Figure 7. Ocular tilt responses associated with lesions at different locations within the peripheral and central vestibular pathways. The ocular tilt reaction consists of a head tilt toward the side of the lower eye, a skew deviation with one eye higher than the other, and counterroll (torsion) of both eyes, with the top poles rolling toward the side of the lower eye. The

ipsilateral eye is down with lesions of the labyrinth and pontomedtillary regions but it is up with lesions in the pontomesencephalic

region.

REFERENCES

1.      Baloh RW . Semicircular canal dehiscence syndrome: leaks and squeaks can make you dizzy . Neurology . 2004 ; 62 : 684 .

2.      Bárány R . Neue Untersuchungsmethoden, die Beziehungen zwischen Vestibularapparat, Kleinhirn, Grosshirn and Ruckenmark betreffend [in German] . Wien Med Wochenschr . 1910 ; 60 : 2033 .

3.      Bárány R . Physiologie and Pathologie des Bogen Gangsapparates beini Menschen . Vienna : Deuticke ; 1907 .

4.      Cho BL , Scarpace D , Alexander NB . Tests of stepping as indicators of mobility, balance, and fall risk in balance-impaired older adults . J Am Geriatr Soc . 2004 ; 52 ( 7 ): 1168 .

5.      Fukuda T . The stepping test: two phases of the labyrinthine reflex . Acta Otolaryngol (Stockh) . 1959 ; 50 : 95 .

6.      Hain TC , Ostrowski VB . Limits of normal for pressure sensitivity in the fi stula test . Audiol Neurootol . 1997 ; 2(6) : 384 .

7.      Hennebert C . A new syndrome in hereditary syphilis of the labyrinth . Presse Med Belg Brnx, 1911 ; 63 : 407 .

8.      Nadol JB . Positive Hennebert’s sign in Meniere’s disease . Arch Otolaryngol . 1977 ; 103 : 524 .

9.      Romberg MH . Lehrbuch der Nervenkrankheiten des Menschen . Berlin, Germany : A Dunker ; 1846 .

10.  Unterberger S . Neue objectiv registrierbare Vestibularis — Korperdrehreaktion, erhalten dutch Freten auf der Stelle: Der “Tretversuch“[in German] . Arch Ohr Nas Kehlk Heilk . 1938 ; 145 : 478 .