The standard electronystagmography/videonystagmography (ENG/VNG) exam, first described 80 years ago, has been around for about 60 years. The recording techniques have improved, but the tests are the same. Our understanding of vestibular function and methods to evaluate the vestibular ocular reflex (VOR) also have improved, but our profession still relies primarily on VNG testing to determine vestibular function. Let’s take a critical look at this standard of care.
The Test Components
The four main components of VNG testing include the following: (1) examination for gaze and spontaneous nystagmus, with and without visual fixation, (2) oculo-motor assessment (including saccade, smooth pursuit, and optokinetic tracking), (3) positional testing, and (4) caloric testing. This article focuses primarily on the role of, and alternatives to, caloric testing. But, first, let’s briefly consider the other test components.
Gaze and Spontaneous Nystagmus
The ability to view for nystagmus under infrared assisted VNG goggles is a great advantage over any other technique. Nystagmus associated with a persistent labyrinthine asymmetry typical of vestibular neuritis, labyrinthitis, or a Meniere’s episode is often not visible through direct viewing due to suppression by visual fixation.
Several alternatives to suppress visual fixation have been suggested, but none are nearly as effective as placing the patient in total darkness with eyes open. Fresnel lenses have been used as a substitute, but Guidetti et al (2006) demonstrated that only about 40 percent of patients with nystagmus visible under VNG goggles had visible nystagmus with Fresnel lenses. Other techniques have been suggested, such as shining a penlight in the eye or having the patient stare at a large piece of paper, but these are not as effective at eliminating visual fixation as VNG goggles (Newman-Toker et al, 2009).
The primary purpose of performing the oculo-motor assessment is to examine motor control of eye movements modulated by the cerebellum. Historically, this information was used to make judgments about cerebellar and brainstem function. Abnormalities were often vaguely described as “central” dysfunction. This test battery was in place long before magnetic resonance imaging (MRI) existed. Today, it is unlikely that a diagnosis of cerebellar, brainstem, or “central” disorder would be made without the benefit of MRI. Still, oculomotor testing plays a role in screening for cerebellar disorders and identifies eye-movement abnormalities that might affect the interpretation of vestibular tests.
Benign paroxysmal positional vertigo (BPPV) is the most common peripheral vestibular disorder (Battacharyya et al, 2017) and, as such, is the most common cause of positional nystagmus. The main benefit of the VNG exam as it relates to BPPV is that Dix-Hallpike and Supine Roll testing are part of the test battery.
While one does not necessarily need the video goggles to make the diagnosis of BPPV, one must complete positional testing to reach a diagnosis of BPPV. Although positional testing under video goggles offers great benefit for replay and review of nystagmus associated with BPPV, this nystagmus is typically visible to the naked eye and does not require VNG goggles in every case.
Some non-BPPV positional nystagmus is only visible with goggles for the same reason some spontaneous and gaze nystagmus is only visible with goggles. It is also beneficial to be able to review, replay, and document other forms of positional nystagmus.
These recordings are useful for (1) review, as the examiner is often attending to the patient in real time at the expense of focusing on the transient nystagmus, (2) replay for patient education, and (3) documentation for second-opinion and billing purposes.
The caloric test has long been used to explore labyrinthine asymmetry. The idea being that, if we stimulate each ear equally, we should get symmetric nystagmus responses to the stimulation. If the responses are asymmetric, that may explain the dizziness symptom that brought the patient to us in the first place. To make this assumption, one must have symmetric stimulation and symmetric anatomy.
Past studies of repeated irrigations on the same subjects demonstrated variations in responses of up to 24 percent between ears (Proctor et al 1986). Variations in temporal bone anatomy can also impact the symmetry of caloric testing, even with normal labyrinthine function (Carpenter et al, 2018).
The vestibular ocular reflex (VOR) works across a range of head speeds in multiple planes. Making judgments about overall VOR function based on caloric results has been proven incorrect. Goebel and Rowdon (1992) determined that two-thirds of patients with absent caloric responses bilaterally have normal VOR function at 0.5 Hz. Of those patients with a clinically significant unilateral caloric hypofunction, 95 percent have normal VOR function at 0.5 Hz. The caloric test provides no information about VOR function in the range of normal head speeds and no information about the functional impact or state of compensation of any caloric abnormality.
The VOR doesn’t have to be very efficient for speeds below 0.5 Hz to provide visual stability, as there are alternative gaze-stabilization strategies available at these slower speeds. At speeds below 0.5 Hz, the VOR is augmented by voluntary ocular tracking, so gain is progressively lower with slower head movements.
At higher speeds, only a functioning VOR allows for gaze stabilization with head movement. With higher speeds of head movement, the gain of the VOR is near 1.0 (otherwise known as “unity gain”), where the velocity of eye movement is equal and opposite of the head movement (see FIGURE 1).
Audiologists are familiar with the unit-to-measure frequency, Hertz or Kilohertz (KHz), from completing hearing evaluations. We know that a 5 KHz sound is a sound wave oscillating at 5,000 cycles per second. In the world of vestibular-function testing, we are most commonly discussing the range between 0.5 Hz and 5 Hz. A 1 Hz head movement would equate to one 360-degree rotation in one second.
For the purposes of describing VOR function (which in real life involves back-and-forth head movement, as opposed to continuous rotation), we will discuss Hz as sinusoidal motion with one cycle including a head movement from center to right shoulder, then smoothly to left shoulder, then back to center. A 1 Hz cycle would take one second to complete. A 0.5 Hz cycle would take two seconds to complete.
A 5 Hz cycle would involve moving the head from center to right shoulder, to left shoulder, back to center, 5 times in one second. This may seem very fast, but this is a speed of head movement frequently encountered during daily activities. Viewing the graph below (FIGURE 2), considering the recording window time to be one second, the top cycle is 0.5 Hz (two cycles), while the bottom tracing is approximately 7 Hz.
The caloric test stimulates the labyrinth through temperature change, altering the density of the inner ear endolymph, which causes fluid movement and a temporary asymmetry in labyrinthine activity. This alters the activity at the level of the vestibular nuclei, creating a sensation of angular movement, analogous to a head movement of around .003 Hz. That would mean a cycle as described above would take more than 240 seconds (4 minutes). This would be considered a non-physiologic speed, as the VOR is not really efficient or needed at this exceedingly low speed.
The graph by Dumas et al (2017) (FIGURE 3) provides a visual of the natural range of head movements, the normal VOR gain range at each frequency, and the frequency range associated with various vestibular function tests. Tests such as rotational chair and active head-rotation tests such as the vestibular autorotation test (VAT) and the video head impulse test (vHIT) provide gain, phase, and symmetry information regarding the current functional level of the VOR.
Rotational chair, while being a very effective test of VOR function, suffers from high equipment cost and marginal reimbursement. Active head rotation, while having lower equipment cost and requiring little test time, suffers from poor test-retest reliability. In contrast, vHIT has several benefits including lower equipment cost, minimal test time, and low stimulation for the patient, which makes associated nausea less likely and provides reliable lateralizing information regarding the high-frequency VOR. Currently, vHIT does not have a code or value assigned for billing purposes and the patient can be directly billed market value.
Several articles have been written regarding the sensitivity of vHIT as a substitute for caloric testing. This is not necessarily helpful because the two tests are not trying to answer the same question. Calorics examine the low-frequency VOR, while vHIT examines the high-frequency VOR. It is important to note that a patient can have normal VOR function for one frequency range and abnormal function for another frequency range.
Sometimes this disagreement can even be beneficial in diagnoses. It has been reported that the pattern of a unilaterally abnormal caloric response and normal ipsilateral vHIT is frequently found in patients with Menieres disease (McCaslin et al, 2014; Hannigan et al, 2019). This is an observation that we have made clinically, as well. vHIT is more relevant than caloric testing in determining functional impact of a vestibular disorder.
Skull vibration is a fast, reliable method of identifying significant labyrinthine asymmetry (Dumas et al, 2017). The premise is that vibration induces global, therefore symmetric, bilateral vestibular stimulation. If both labyrinths register the stimulation equally, then nystagmus is not elicited. If a significant labyrinthine asymmetry exists, nystagmus is generated, beating toward the ear registering greater excitability.
In a labyrinthine hypofunction such as encountered in vestibular neuritis/labyrinthitis or vestibular schwannoma, the nystagmus beats away from the lesion. In lesions associated with increased labyrinthine response, such as superior semicircular canal dehiscence syndrome (SCDS), the nystagmus beats toward the lesion.
Skull vibration involves low equipment cost, as it can be completed with a simple handheld massager. One does need access to VNG goggles, though, for optimal testing. Skull vibration is very fast, repeatable, and not affected by state of compensation. Even if caloric testing is performed, skull vibration can provide information that may allow a cross-check of caloric findings, potentially reducing the number of irrigations required. At this time, there is not a billable insurance code for skull vibration and the patient can be billed directly at market value.
Caloric Testing as the Gold Standard?
Both vHIT and skull vibration have been examined for sensitivity and specificity using caloric results as the gold standard or benchmark. A gold standard is most commonly agreed upon as a measure or technique that is widely used with high accuracy and sensitivity. Caloric testing may meet the first standard in that it is, or at least has been, widely used. We disagree that it is the most accurate or sensitive test possible to identify patients with symptomatic vestibular dysfunction.
In 2010, Medicare reimbursed $98.80 for bilateral bi-thermal caloric testing. Current reimbursement for the same test is around $41.00, an almost 60 percent reduction in payment.
“Caloric testing in the USA is dying out,” according to the website dizzinessandbalance.com, run by Dr. Tim Hain. The site’s section on VNG testing begins with: “The reason is that doing caloric testing is generally a financial loss for the outpatient medical facility, because it takes a long time (often more than an hour), it requires an expensive piece of machinery, it requires a highly trained individual to do the test, and because many large insurance companies and Medicare pay very little for the caloric test. If you are fortunate enough to find an outpatient facility that still does good quality calorics, you are lucky!”
Caloric testing has the potential to point some clinicians in the wrong direction in determining the exact source of symptoms. For example, a patient complaining of brief positional vertigo, even with a negative Dix-Hallpike test, is more likely suffering from BPPV with a negative examination than from a chronic labyrinthine asymmetry, despite what caloric test results show.
A patient with transient postural lightheadedness and imbalance is more likely suffering from orthostatic hypotension than from a mild caloric hypofunction that was detected. Of course, an existing vestibular hypofunction has the potential to exacerbate any imbalance or dizziness symptoms associated with postural lightheadedness and BPPV may be developed secondary to a pre-existing vestibular injury. This is where a thorough case history and a review of pertinent medical conditions and medications are necessary.
Calorics are the most unpleasant, time consuming part of the evaluation for all involved. Technology has changed. We have better alternatives that are faster and easier for examiner and patient and have more functional relevance.
Calorics are still important for some patients but, in our opinion, dependent on symptoms, you can safely skip them with many patients.