Monitoring Auditory Evoked Potentials

Møller, Aage R.

doi:10.1007/978-1-4419-7436-5_7

Aage R. Møller PhD (DMedSci)²

2316 Accesses

Abstract

Monitoring of the auditory system makes use of subcortical evoked potentials [Auditory Brainstem Responses (ABR) and CAP recorded from the exposed auditory nerve]. These modalities of evoked potentials are not affected by commonly used anesthetics.

Download chapter PDF

Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

Monitoring of the auditory system makes use of subcortical evoked potentials [Auditory Brainstem Responses (ABR) and CAP recorded from the exposed auditory nerve]. These modalities of evoked potentials are not affected by commonly used anesthetics.

One purpose of monitoring auditory evoked potentials is to reduce the risk of injury to the eighth cranial nerve (CN VIII), which is at risk of being injured by surgical manipulations in microvascular decompression (MVD) operations to relieve trigeminal neuralgia (TGN), hemifacial spasm (HFS), glossopharyngeal neuralgia (GPN) (1, 2), and in connection with MVD operations of the eighth nerve in patients with tinnitus and disabling positional vertigo (DPV) (3). Preservation of auditory function during the removal of small vestibular schwannoma has recently improved due to advancements in operative techniques and through the introduction of intraoperative neurophysiological monitoring of the auditory nerve (4–9).

ABR were some of the earliest sensory evoked potentials to be used intraoperatively for the purpose of reducing intraoperative injuries to the auditory nerve (1, 10). In operations to remove vestibular schwannoma, recordings of ABR have been supplemented by recording CAP from the exposed CN VIII (2, 4–7) and from recording evoked potentials from the vicinity of the cochlear nucleus (8, 11, 12) for monitoring of the integrity of the function of the auditory nerve.

Only the auditory part of CN VIII can be monitored, but it has been shown that the vestibular part of CN VIII can be injured in MVD operations and can produce symptoms and signs that indicate insult to the balance system (13). The advantages and disadvantages of different methods for monitoring the integrity of the auditory nerve are discussed, and different ways to optimize such recordings are described.

Recording from the vicinity of the cochlea (electrocochleography, ECoG) has been described as a technique for monitoring hearing in operations on vestibular schwannoma (14, 15). However, since the risk of damage to the auditory nerve in most operations affects its intracranial portion, and the ECoG potentials only reflect the CAP of the distal part of the nerve, the usefulness of monitoring ECoG is limited. Changes in the ECoG potentials, however, indicate impairment of blood supply to the ear. If that is caused by permanent damage to the labyrinthine artery, it is normally not reversible and monitoring cannot prevent permanent loss of hearing.

The choice of acoustic stimuli and how they are presented, as well as the hearing status of the patient, can influence the amplitude, latency and waveform of the recorded potentials (ABR or CAP). It is, therefore important to consider these factors in the interpretation of the results of intraoperative monitoring of auditory evoked potentials. Thus, all patients in whom intraoperative monitoring of auditory evoked potentials is to be performed should have hearing tests performed preoperatively. Included in such tests should, at the very least, include pure tone audiometry, determination of speech discrimination (using recorded speech material),and ABR. It is also preferable to include testing of the acoustic middle ear reflex.

Such preoperative tests are also a prerequisite in order to quantitatively evaluate a change in hearing status that may occur as a result of an intraoperative injury to the auditory nerve. These tests also assess the value of intraoperative monitoring of auditory evoked potentials and the value of any modification in the usual surgical methods that may be made in an attempt to improve hearing preservation (see Chap. 19).

This chapter discusses the practical aspects of hearing preservation in various types of operations using recordings of ABR or CAP directly from the auditory nerve or the vicinity of the cochlear nucleus. Recordings of ABR have been used to detect effects on the brainstem from surgical manipulations during operations on large vestibular schwannoma and on other types of masses that may occur in the cerebellopontine angle (CPA) (6, 16, 17), as well as on tumors or other space-occupying lesions in the region of the fourth ventricle.

Auditory Brainstem Responses

The ABR was described in Chap. 5. The technique used in recording ABR for intraoperative monitoring is similar to that used clinically to obtain ABR for diagnostic purposes. However, when recording ABR intraoperatively, several modifications in this technique are necessary because of the special environment of the operating room and because it is important to be able to obtain an interpretable record in as short a time as possible.

It is mainly changes in the latencies of specific components of the recorded evoked potentials (ABR or CAP from the auditory nerve or the cochlear nucleus) that are used as indications of injuries to the auditory nerve, but changes in amplitude of the recorded evoked potentials are also valuable signs of surgically induced injuries (18).

Changes in CAP recorded from the auditory nerve provide direct information about changes in the function of the auditory nerve, while interpretation of the intraoperatively recorded ABR is more complex.

Since the purpose of intraoperative monitoring of ABR is to detect changes that occur in the patient’s auditory system during an operation, the recordings that are made during the operation must be compared with a baseline recording obtained in the same patient before the operation began rather than with a standard ABR recording as is made when ABR are used for clinical diagnostics.

How to Obtain an Interpretable Record in the Shortest Possible Time?

The ABR obtained during an operation must be interpreted immediately after they are completed so that changes in the ABR can be identified with the shortest possible delay, and the information can be conveyed promptly to the surgeon. The criteria for obtaining a response as quickly as possible have similarities with those for SSEP, but the amplitudes of the ABR are much smaller than those of SSEP.

Because the ABR have much smaller amplitudes than the background of noise in the operating room (consisting of ongoing biological activity such as spontaneous activity from the brain, muscle activity, and electrical interference), many responses must be added (averaged) to obtain an interpretable record. The time it takes to obtain an interpretable record, therefore, depends on the amplitude of the ABR in relation to the background noise (the signal-to-noise ratio) and how many responses can be added per unit of time, based on the repetition rate of the stimuli. The most important factors for obtaining an interpretable record in the shortest possible time are:

1.
Stimulus intensity
2.
Stimulus repetition rate
3.
Electrode placement
4.
Electrical and other interferences
5.
Filtering of recorded potentials
6.
Quality control that does not add time to data collection.

Stimulus Intensity.

The stimulus intensity should be adequately high, without imposing a risk of causing noise-induced hearing loss (NIHL) so that the amplitude of the recorded ABR is as high as possible. Clicks at an intensity of 105 dB peak equivalent sound pressure level (PeSPL) have been used for intraoperative monitoring for many years without problems. This intensity corresponds to ∼65 dB hearing level (HL) (HL – dB above the average threshold of hearing in individuals with normal hearing, when click sounds are presented at a rate of 20/s).

Stimulus Repetition Rate.

When the stimulus repetition rate is increased, the number of responses that can be collected within a certain period of time increases. If the amplitude of the responses were independent on the repetition rate, then the time it would take to obtain an interpretable record would be inversely proportional to the repetition rate, thus a doubling of the repetition rate would shorten that time by a factor of two. However, this is only the case below a certain repetition rate because the amplitude of the peaks decreases with increasing repetition rate above a certain repetition rate and diminishes the gain of increasing the repetition rate. The decrease in amplitude that occurs when the repetition rate is increased is minimal at low repetition rates, but it accelerates with increasing repetition rate (Fig. 7.1A). There are only small changes in the ABR when stimulus repetition rates are increased from a few stimuli per second up to 20 stimuli per second. At a certain repetition rate, the reduction in amplitude of the recorded potentials becomes so great that it outweighs the gain from producing more responses per unit time (Fig. 7.1B). This is the repetition rate that provides an interpretable record in the shortest possible time. If the repetition rate is increased beyond that rate, it will take a longer time to obtain an interpretable record. No data are available for the optimal stimulus repetition rate for ABR.

The relationship between the repetition rate of the stimulation and the amplitude of the individual peaks of the ABR depends on the individual’s age and hearing loss, and increasing the stimulus repetition rate affects the different peaks differently. Peaks I–III are much more affected by an increased repetition rate than peak V, which is the most robust of the peaks of the ABR with regard to high repetition rate of the stimulus (19).

Hearing loss of cochlear origin does not seem to affect the way that the amplitude of the ABR peaks decrease with increasing repetition rate of click stimuli, but hearing loss of retrocochlear origin, such as caused by an injury to the auditory nerve, affects how the amplitude of peak V decreases with increasing repetition rate of the stimulus. The product of the amplitude of peak V and the repetition rate of the click stimuli in individuals with hearing loss of retrocochlear origin (presumably from injury to the auditory nerve) nearly reaches a plateau somewhere above 40 pps (20) (Fig. 7.1B). Other investigators (21) obtained similar results. On the basis of these results, it seems advantageous to use repetition rates of at least 50 pps, and perhaps as high as 70 pps. That is much higher than the commonly used repetition rate (10–20 pps) (20) (Fig. 7.1). (Because the time required to obtain an interpretable record when recording ABR in the clinic is not important, most clinical recordings of ABR employ a low repetition rate of 10 to 20 pps).

Since it is not completely known how disease processes that affect the ear and the auditory nerve can affect the relationship between stimulus repetition rate and the amplitudes of the various peaks, it may not be advisable to use repetition rates higher than 50 pps. When the repetition rate is increased, caution should be exercised because the risk of (noise-induced) hearing loss from the sound increases accordingly, and it may not be advisable to use repetition rates higher than 40 pps if an intensity of 105 dB PeSPL is being used.

The fact that the latencies of the peaks of the ABR increase with increasing stimulus repetition rate is not important for the selection of the stimulus repetition rate for ABR in the operating room because in the operating room, the patient’s own ABR serve as the reference (baseline), provided that the same repetition rate is used for monitoring as was used for obtaining the baseline recording.

Sound Delivery.

Several kinds of insert earphones are suitable for use in the operating room to deliver sound stimuli for recording ABR. The miniature earphones used with, for instance, typical MP3 players, have a broad frequency response and can easily be fitted into the ear of a patient in the operating room. This author has used similar earphones (Radio Shack^{Footnote 1}) routinely in the operating room for many years. The earphones are normally driven by rectangular waves of 100-microsecond (μs) duration. These earphones deliver a narrow sound impulse and have a maximal sound output of approximately 110 dB PeSPL and deliver clicks of 105 dB PeSPL without any noticeable differences in amplitudes or waveforms of rarefaction and condensation clicks (corresponding to ∼65 dB HL when presented at a rate of 20 pps). The frequency spectrum of the clicks that are generated by these earphones is relatively flat over a large range of frequencies (100–7,000 Hz ± 8 dB) with a broad peak around 5 kHz when measured at the entrance of the ear canal. The sound spectrum is the product of the frequency transfer function of the earphone and the spectrum of the electrical impulses used to drive the earphone. When using a square wave of 100 μs duration, there is a dip in the spectrum of the sound at 10 kHz because of the spectrum of the electrical input to the earphone. The spectrum of a square wave of 100 μs duration has a cutoff at 8,000 Hz (6 dB), and its energy is zero at 10 and 20 kHz causing dips in the spectrum of the sound at these two frequencies (22). In fact, the commonly used duration of the rectangular impulses of 100 μs is not ideal; both longer and shorter durations are more suited for driving the sound generators used for eliciting ABR (see (22)).

When such a miniature stereo earphone is placed in the ear of a patient, it should be placed so that its sound-radiating (flat) surface faces the ear canal and that the earphone does not just rest in the pinna. This is particularly important to consider when such an earphone is placed in the ear of patients who have large outer ears (pinna), which is often the case in elderly men. The earphone must be carefully secured in place with several layers of a good-quality plastic adhesive tape (e.g., 3 M Company^{Footnote 2} Blenderm^R) in such a way that fluid cannot reach the earphone just in case the area around the ear should get wet. The cord to the earphone must be secured with adhesive tape to the side of the patient’s face and to the head-holder (or operating table) so that the earphone is not accidentally dislodged from the ear if the cable is accidentally pulled.

Some of the modern insert earphones usually have the transducer connected to the ear by means of a plastic tube of various lengths. When driven by the standard rectangular wave of 100 μs duration, some earphones deliver a sound with a relatively flat spectrum up to ∼6 kHz, which is similar to the spectrum delivered by the earphones used in audiometry and those often used in clinical ABR testing. The fact that insert earphones deliver sound through a long (plastic) tube results in a delay between the delivery of the electrical impulse that drives the earphone and the arrival of the sound at the ear. Sound travels at a speed of about 340 m/s, corresponding to a delay of 1 ms/34 cm, thus the delay is slightly less than 1 ms for each foot of tubing. A delay of 1 ms makes the (electrical) stimulus artifact appear 1 ms ahead of the sound’s arrival at the ear and thus, reduces interference from the stimulus artifact with the ABR (see also Chap. 18 regarding how to reduce stimulus artifacts).

Electrode Placement.

The electrodes used for recording ABR should be placed so that the amplitude of the recorded potentials is as high as possible and so that the components of the ABR that are of interest appear as clearly as possible. The traditional way of recording ABR is by connecting one of the two inputs of a differential amplifier to an electrode placed on the vertex and connecting the other input to an electrode placed on the ipsilateral earlobe or the ipsilateral mastoid.

As mentioned above, this placement of recording electrodes for ABR recordings is not ideal for obtaining the largest possible potentials. Instead, recordings should be made in accordance with the orientations of the dipoles that represent the different components of the ABR. Scherg and von Cramon (23) showed that the generation of the different components of the ABR could be synthesized by six dipoles that were approximately located in the coronal plane (a vertical plane that is perpendicular to the saggital plane). Dipole of peaks I and III are approximately oriented horizontally, and peak V is nearly vertically oriented (Fig. 7.2). The negative troughs that follow peak I and peak III are oriented slightly differently. This means that electrodes placed in the horizontal plane record peaks I and III optimally, and peak V is best recorded by electrodes placed in the vertical plane. Using two separate recording channels, one recording differentially between electrodes placed at the vertex and on the dorsal upper neck (a noncephalic reference) and the other recording differentially from electrodes placed on the two earlobes, thus record peak V optimally and peaks I–III optimally. This way of recording ABR provides a record in which peak V appears more distinctly in the recording from the vertex–neck placement of the electrode, and peaks I and III are better represented in the recording from the earlobes than that which can be seen in the traditional way of recording ABR from electrodes placed at the vertex and on the ipsilateral earlobe.

Recording in two independent channels offers two alternative ways to detect changes in auditory function during an operation, and it makes it possible to continue monitoring using only one channel if one of the electrodes should malfunction during an operation.

If the recordings in one of these channels change noticeably, the surgeon should be informed. This does not need to be an alarm, as it is now popular to define a certain change in the recorded potentials as a warning. Any change that is larger than the normal small fluctuations should be reported to the surgeon because such changes mean that some structure has been affected.

The equivalent dipoles shown in Fig. 7.2 were derived from recordings in three channels from three pairs of electrodes placed orthogonally on the scalp (24–26). Each pair of electrodes is connected to the two inputs of three independent differential amplifiers. The recorded potentials are then plotted as a function of each other to form a three dimensional display with time as a parameter.

An example of recordings of the ABR in three orthogonal planes is shown in Fig. 7.3A. When combined, such recordings are known as three-channel Lissajous’ trajectory as shown in Fig. 7.4B. Such recordings, which provide a complete description of the ABR, have been used to determine the neural generators of the ABR (27).

Such recordings provide information about the anatomical location of the neural generators of the various components of the ABR in the head because they take into account the orientation of the different dipoles. There is, however, some uncertainty regarding the interpretation of the potentials when they are recorded in this way. This type of recording is not commonly used in intraoperative monitoring, but has been used for research purposes in the operating room (28) and may find use in intraoperative monitoring in the future.

Types of Electrodes.

When ABR are recorded for clinical diagnostic purposes, it is convenient to use surface electrodes to record the responses, but in the operating room, needle electrodes or wire hook electrodes are more suitable for several reasons. When held in place with a good-quality plastic adhesive tape (for instance, 3M, Blenderm^R2), needle electrodes or wire electrodes provide a more stable recording over a longer period of time than do surface electrodes. Platinum subdermal electrodes (or disposable electrodes that are available from numerous sources) are suitable. The same is the case for wire hook electrodes. Inserting needle electrodes or wire hook electrodes are usually applied in the operating room after the patient is anesthetized, and there is no discomfort associated with placing such needles on anesthetized patients. At the end of the operation, the electrodes should be taken out before the patient is awake.

All precautions should be taken to avoid failure of any recording electrodes during an operation. It is, thus, important that the electrodes be inserted properly and secured well to reduce the risk that they become dislodged should the electrode wires be accidentally pulled or should the area where the electrodes are placed be manipulated during the operation. The electrode placed on the vertex for recording ABR must be inserted deep in the tissue, and the electrode wire must be drawn toward the forehead and placed under the hair as close to the skin as possible and then secured to the forehead with adhesive tape. When recording from a person with much hair, the movements of the drape can make the hair move, and if the electrode wire is resting on top of the hair, it too moves and results in a noisy recording or a dislodged electrode.

For these reasons, surface electrodes are not suitable for ABR recordings. In operations in which skin incisions are made near the earlobe, the earlobe electrode may be pulled out if it is not sufficiently secured with adhesive tape or with sutures.

Processing of Recorded ABR

Because mainly changes in the latency of peak V (and to some extent of peak III) are used in connection with intraoperative monitoring, it is important that these peaks appear as clearly as possible in the recordings. The purpose of processing recorded ABR is, therefore, to enhance these peaks (III and V) so they can be clearly identified and their latency can be measured. This can be performed by two methods: (1) averaging the responses to a sufficient number of stimuli, and (2) suitable filtering of the responses. Signal averaging increases signal-to-noise ratio (SNR) by adding responses to many stimuli. The purpose of filtering is to attenuate signals that are not wanted (noise) and to enhance features of the recorded potentials that are important for their interpretation. The latter can be performed either at the same time that the responses are recorded using analog filters or after the responses have been averaged using computer programs using digital filters (see Chap. 18).

Filtering of Recorded Potentials.

There are several reasons for filtering recorded potentials such as ABR. As is discussed in Chap. 18, high-frequency energy must be attenuated before the recorded signals are sampled and converted to a stream of digits. Another reason for filtering is to suppress background noise as much as possible. A third reason is to enhance important factors of recorded potentials.

Recorded ABR are, therefore, always subjected to some form of spectral filtering; analog filtering is used before the recorded responses are digitized for signal averaging to avoid aliasing (see Chap. 18). After being converted to a digital form, the recorded potentials may be filtered by digital filters, which are computer programs that perform the filtering. Digital filters and their use in filtering evoked potentials are described in Chap. 18. Digital filters have many advantages over analog filters for attenuating noise and for enhancing the waveform of evoked potentials, such as ABR, as illustrated in Fig. 7.4.

The purpose of processing the recorded ABR is to obtain a record that is as clear as possible and to enhance features that are of interest, an example of which is seen in Fig. 7.5. The techniques that are suitable for processing ABR are similar to commonly utilized methods for processing other evoked potentials (for details see Chap. 18).

In the recordings illustrated in Fig. 7.4, analog filters were set at rather “open” values; 10 Hz high-pass and 3-kHz (kilohertz) low-pass, and the slope of the high-pass filter was 6 dB/octave and that of the low-pass filter was 24 dB/octave. The digital filters were zero-phase finite impulse response filters as described in Chap. 18 (29, 30). The TRI 10 filter only smoothes the recordings, but the filters used for the lower two traces enhance the peaks, and the ABR shown in the two lower graphs have a much clearer definition of the peaks than the ABR that were only subjected to analog filtering. (The use of zero-phase finite impulse digital filters is discussed later in this book, Chap. 18).

Quality Control.

Quality control of recorded potentials is important. In the clinic, quality control is performed by response replication. This is not a suitable method for intraoperative monitoring because having to make two recordings extends the time to get an interpretable recording. Methods that do not require repeating the response, and thus, do not take any additional time, are described in Chap. 18.

Display of ABR in the Operating Room

When monitoring ABR in the operating room, several tracings should be displayed, namely, the digitally filtered, averaged ABR recorded on two channels. One is recorded differentially between electrodes placed on the vertex and the dorsal neck, and the other channel should be recorded differentially between the two earlobes. The filtered ABR should be superimposed on a baseline recording on both of these channels. It is also important to have a display of the direct output of the amplifiers of the ABR in order to be able to evaluate background noise. If the output of the amplifier is not monitored, suddenly occurring interference would only be detected by an increase in the number of rejected responses, and that does not provide information about the kind of interference. Only by continuously observing the output from the amplifier is it possible to identify the source of interference (see Chap. 17 for details).

Recording of Near-Field Potentials

Recordings of near-field potentials from structures of the ascending auditory pathways in humans were first made for research purposes (11, 31–37), but have later found practical application in intraoperative monitoring, particularly for reducing the risk of injures to the auditory nerve (4, 5, 7, 38). Recordings from the exposed auditory nerve or from the surface of the cochlear nucleus are valuable in monitoring neural conduction in the auditory nerve (9).

Direct Recording from the Eighth Cranial Nerve

Recordings of CAP from the auditory nerve in such operations can be performed by placing an electrode on the exposed CN VIII. In response to transient sounds (clicks or tone bursts), such recordings yield CAP with amplitudes of a few microvolts in patients with normal hearing (31, 32, 37, 39). These potentials can, therefore, be displayed directly or after only a few responses have been averaged. This method provides a much more rapid way to detect injuries to the auditory nerve than monitoring ABR in MVD operations to move blood vessels off different cranial nerves in disorders, such as hemifacial spasm, trigeminal neuralgia, tinnitus disability, and in monitoring of operations to remove vestibular schwannoma (4, 5, 7, 31).

A fine, malleable, single-strand, Teflon-insulated silver wire (Medwire Corporation^{Footnote 3} Type Ag 7/40 T) (31) has been used by the author for many years. About 2 mm of the insulation is removed from the tip of this wire; the bare wire is then bent, and a small piece of cotton is sutured to the tip using a 5-0 silk suture. The cotton is then trimmed using microscissors to produce the finished electrode shown in Fig. 7.5. It is important that the cotton wick is securely sutured to the wire, since the electrode is to be placed on the exposed eighth nerve and losing a piece of cotton in the CPA can have serious consequences. Shredded Teflon^® offers the same advantage as cotton but creates a less adverse reaction if accidentally lost intracranially. After the cotton, wick is sutured to the silver wire, it is soldered to a PVC-insulated and electrostatically shielded wire that connects the electrode to the input of the amplifier (electrode box).

In operations in the CPA, the recording electrode wire is tucked under one of the sutures that holds the dura open. In addition, the electrode wire is clamped to the drape near the incision to secure it in place.

The wire from the recording electrode should be connected to the inverting inputs of a differential amplifier so that a negative potential causes an upward deflection on the screen. The shield of the wire should be grounded to the iso-ground of the amplifier. The reference electrode for the intracranial recordings can be placed in the opposite earlobe.

The Anatomy of CN VIII.

CN VIII comprises the vestibular nerve and the auditory (or cochlear) nerve. The arrangement of the different components of the eighth nerve is seen in a cross-sectional view, illustrated in Fig. 7.6, which shows the eighth cranial nerve inside the internal auditory meatus. The auditory nerve is located on the caudal side of the eighth nerve near the brainstem and anterior ventral to the eighth nerve near the porus acousticus. The nerve is rotated as it passes from the ear to the brain stem (Fig. 7.6 and 7.7).

The CAP that can be recorded from the auditory nerve in a patient with normal or near normal hearing – with the recording electrode placed on the nerve near the porus acousticus – has a triphasic waveform (Fig. 7.8A). An initial (small) positive peak is followed by a large negative peak, which in turn is followed by another small, positive peak. This is what may be expected when recording from a long nerve using a monopolar electrode (see Chap. 3). The amplitude and the waveform of the CAP depend on the stimulus intensity (Fig. 7.9A) and on the placement of the electrode along the auditory nerve (Fig. 7.9B).

The size of the recorded potentials is largest when the recording electrode is placed on the auditory portion of the eighth nerve, but even when placed on the vestibular portion of the eighth nerve, the recorded potentials (CAP) are normally several microvolts (μV) and large enough to be visible directly on a computer screen (or after averaging only a few responses). The reason that potentials of such large amplitude can be recorded, even when the electrode is placed on the vestibular portion of the eighth nerve, is that the vestibular nerve is a good conductor of electrical current.

The waveform of the normal CAP is essentially the same when using 2-kHz tone bursts as stimuli as when using clicks, but the changes in the responses as a result of pathologies affecting the ear or the auditory nerve may be different for click sounds than for tone bursts. The waveform of the CAP when recorded in the same way in patients with hearing loss (Fig. 7.9) may deviate noticeably from the waveform shown in Fig. 7.9.

Recording from the Vicinity of the Cochlear Nucleus

The value of monitoring directly recorded evoked potentials from the exposed auditory nerve is well documented as shown above. However, the difficulties in placing the electrode in the correct position on the eighth nerve are obstacles to the routine use of such directly recorded evoked potentials. The recording electrode must be placed proximal to the location on the nerve where it is at risk of being injured, and it may be difficult to keep the recording electrode in the correct position at times during an operation. These problems hamper the general use of recording directly from the auditory nerve.

Recording from the vicinity of the cochlear nucleus (11, 12) can overcome many of the practical difficulties associated with recording directly from the exposed eighth nerve, and it has similar advantages as recording CAP directly from the eighth nerve (8, 9). The cochlear nucleus forms the floor of the lateral recess of the fourth ventricle (8, 40), and recording from the vicinity of the cochlear nucleus can be performed by placing a recording electrode in the lateral recess of the fourth ventricle (8, 9) (Fig. 7.10A).

The same type of wick electrode (Fig. 7.5A) as used to record from the exposed eighth nerve can be used for recording from the surface of the cochlear nucleus. The opening of the lateral recess of the fourth ventricle, known as the foramen of Luschka, is found just anterior to the entrance of the CN IX/CN X complex into the brainstem. The foramen of Luschka may be identified by locating the choroid plexus that normally protrudes from the foramen. Elevating the cerebellum over the CN IX/CN X complex provides access to the foramen of Luschka. By following the choroid plexus into the lateral recess of the fourth ventricle, the recording electrode may be placed deep into the lateral recess (8). The wire of the recording electrode should be tucked under the sutures that hold the dura open so that it cannot be easily moved during the operation (Fig. 7.10A). The recording electrode should be connected to the inverting input of the amplifier just as it is connected when recording from the exposed CN VIII. The opposite earlobe is a suitable location for the reference electrode for such recordings.

Recorded potentials from the surface of the cochlear nucleus consist of an initial sharp, positive–negative deflection that is generated by the termination of the auditory nerve in the cochlear nucleus. This peak is followed by a slow wave that may last tenths of milliseconds and which has several waves riding upon it (Fig. 7.10B). The waveform of the potentials recorded from the surface of the cochlear nucleus (or its vicinity) resembles that which is normally recorded from a nucleus (see Chap. 2, Fig. 2.7). As seen from Fig. 7.10B and C, preoperative hearing loss affects the waveforms of the potentials that are recorded from the surface of the cochlear nucleus.

Recordings of evoked potentials from the cochlear nucleus or its vicinity represent evoked potentials that are generated by structures located proximal to the auditory nerve. Changes in these potentials are, therefore, good indications of changes in the function of the auditory nerve such as those that may occur when the nerve is being manipulated, such as in MVD operations. Recordings from the cochlear nucleus are, however, perhaps of the greatest importance in connection with the removal of vestibular schwannoma in patients who have useful hearing preoperatively and in whom hearing preservation is being attempted during the removal of the tumor (see page 137).

The fast components of the response from the cochlear nucleus are generated where the auditory nerve terminates in the cochlear nucleus, and the slow components are generated by dendrites. The fast components are, thus, directly associated with neural transmission in the auditory nerve and signal arrival of neural activity in its target neurons. The fast components are, therefore, probably the best indicators of injury to the auditory nerve, and they are the components that should be watched in intraoperative monitoring of the auditory nerve.

Digital filters can be used to enhance the fast peaks of the responses and suppress the slow components (Fig. 7.11). Changes in the stimulus intensity affect the fast (initial) and the (later) slow potentials differently. The amplitude of the main peak of the fast response that occurs with a latency of ∼4 ms decreases rapidly when the stimulus intensity is decreased, while the slow components that dominate the unfiltered response only change minimally with decreasing stimulus intensity (Fig. 7.11). It is not known which of these components, slow or fast, are the best indicators of injury to the auditory nerve, but it seems likely that the fast components (such as the negative peak at 4 ms) would be more sensitive to changes in neural conduction in the auditory nerve than the slow components.

It may sometimes be difficult to place the recording electrode deep in the lateral recess of the fourth ventricle, but it is not necessary to penetrate the foramen of Luschka with the recording electrode to obtain satisfactory recordings; merely placing the recording wick electrode on CN IX and X where they enter the brainstem usually provides a satisfactory recording of the response from the cochlear nucleus. The amplitudes of these potentials may be slightly lower than those recorded from an electrode placed deep in the lateral recess, but the potentials that are recorded from the entrance of CN IX and CN X in the brainstem are usually several microvolts, and can, thus, be interpreted after only a few hundred responses are added. It is easier to place the recording electrode in this location than it is to place it on the eighth nerve, which is an advantage when monitoring operations for vestibular schwannoma.

It is practical to record ABR and the potentials from the lateral recess simultaneously to produce different traces on the display. The same stimuli used to elicit ABR are also suitable to elicit these directly recorded potentials from the surface of the cochlear nucleus.

Interpretation of Changes in Auditory Responses

The changes that occur during an operation should be related to specific manipulations, such as stretching, compressing or heating neural tissue, and the anatomical location of the structures whose functions have changed should be identified to the surgeon. Such events should be documented in the final report of the monitoring together with the recordings of the evoked potentials.

Interpretation of Changes in the ABR

In the operating room, the task is to detect changes in auditory evoked potentials from a baseline recording that is made after the patient is brought to the sleep stage of anesthesia, but before the operation has begun. Traditionally, it has been the latency of specific components (vertex positive peaks) of the ABR that has been used to indicate surgically induced injuries to the auditory nerve. Since peak V of the ABR is the most prominent and most easily identified peak, it seems natural to use changes in the latency of this peak as an indication of injury to the auditory nerve. It has often been assumed that any change in neural conduction of the auditory nerve is equally reflected in the latencies of peak II and any one of the ABR peaks that follows peak II.

It is also of value to observe changes in the amplitude of the components of the ABR. It has been shown that inclusion of changes in the amplitude increases the value of the ABR for detecting changes in the function of the auditory nerve (18, 41).

However, changes in the amplitudes of peak III and peak V are not necessarily the same, and there are, therefore, reasons to monitor changes in the amplitudes of both peak III and peak V. Changes in the function of the auditory nerve may cause a smaller change in the amplitude of peak V than that of peak III. Peak V may, therefore, be less sensitive to injury to the auditory nerve than peak III or the CAP recorded directly from CN VIII or the cochlear nucleus. The amplitude of peak III may be a more reliable (clean) indicator of changes in neural conduction of the auditory nerve than peak V. Often the vertex-negative peak between peak III and peak IV–V complex is prominent, and in such cases, using this vertex-negative peak (valley) is just as suitable for monitoring purposes as peak III.

If the latency of peak V increases, but the latency of peak III remains unchanged, the interval between peaks III and V increases (increased interpeak latency, IPL, III–V). The reason for such a change is most likely altered functions of structures of the ascending auditory pathways that are located rostral to the generators of peak III (mostly the cochlear nucleus). Increased IPL III–V may also be caused by general changes in, for example, cerebral circulation or from changes in oxygenation (causing ischemia), which can have many causes. If this occurs in operations in the CPA, the anesthesiologist should be informed because ischemia may be a result of cardiovascular changes or other changes that the anesthesiologist can correct.

Interpretations of CAP from CN VIII and the Cochlear Nucleus

Changes in the CAP recorded directly from the proximal portion of the auditory nerve as a result of manipulation of CN VIII are more easily interpreted than changes in the ABR. The CAP recorded from CN VIII or the cochlear nucleus is probably also more sensitive to small changes in the function of the auditory nerve than are the ABR. Recording of ABR is, however, the only way to detect injuries to the auditory nerve that may occur before surgical exposure of the eighth nerve. Changes in the ABR may occur during retraction of the cerebellum that can stretch the auditory nerve, or it may be caused by surgical dissection to expose the auditory nerve.

The major advantage of recording directly from the exposed CN VIII or the cochlear nucleus is that changes in neural conduction in the auditory nerve can be detected almost at the moment they occur. The large amplitude of the CAP recorded directly from the auditory nerve allows the CAP to be viewed on a computer screen once a few responses have been averaged, making it possible to accurately identify which steps in an operation caused change in neural conduction in the auditory nerve. The rapid detection of change in neural conduction of the auditory nerve also provides a much better possibility to reverse a surgically induced change in the function of the auditory nerve, thus increasing the effectiveness of intraoperative monitoring. Assessment of neural conduction in the auditory nerve on the basis of changes in ABR takes a much longer time than from inspection of the CAP recorded directly from the auditory nerve or recordings from the cochlear nucleus.

The first CAP that are recorded should be used as a baseline to which successive recorded potentials can be compared. Any deviations in the components from the baseline recording should be regarded as a sign of an effect on neural transmission in the part of the auditory nerve that is located distal to the location of the recording electrode on the nerve. The recording electrode should, therefore, be placed as far proximal on the cochlear–vestibular nerve as possible, or it should be placed on the cochlear nucleus. Recordings from the cochlear nucleus reflect the neural conduction in the entire auditory nerve.

Heating from electrocoagulation can cause changes in the waveform of the CAP recorded from the exposed auditory nerves as illustrated in Figs. 7.12 and 7.13.

The change in the CAP recorded from the auditory nerve that may occur as a result of surgical manipulations or heating is a more or less marked decrease in the amplitude of the main negative peak of the CAP. In addition, an increased latency and increased amplitude of the initial positive wave may occur. The increased amplitude of the initial positive wave (downward deflection in Figs. 7.12 and 7.13) indicates that a conduction block has occurred in many nerve fibers. The recordings shown in Fig. 7.13 illustrate changes that occurred after heating of the auditory nerve by electrocoagulation. Shortly after the eighth nerve was exposed, the recorded CAP had the normal triphasic waveform (Fig. 7.13), but after electrocoagulation of a nearby vein, it gradually changed and became a single positive wave (Fig. 7.13), which indicated that there was nearly total blockage of neural conduction in the auditory nerve.

In order to understand the nature of this kind of injury, the generation of the CAP from a long nerve, when recorded by a monopolar electrode, should be recalled. The initial positive deflection in the CAP is generated by a region of neural depolarization as it approaches the site of the recording electrode, and the negative peak in the CAP is generated when the region of depolarization of auditory nerve fibers passes under the recording electrode (see Chap. 2). The near disappearance of the negative peak (Figs. 7.12 and 7.13) can be explained by the region of depolarization never reaching the location on the nerve where the recording electrode is placed. The amplitude of the initial positive peak in the CAP – which is generated when the region of depolarization of nerve fibers approaches the recording electrode – is normally decreased because the negative peak that normally follows pulls up the positive peak. When the amplitude of the negative peak decreases, this “pull” of the negative peak on the positive peak upward decreases, and therefore, the positive peak appears to have become larger in amplitude.

Examples of changes in the CAP caused by the retraction of the cerebellum are seen in Fig. 7.14. The slight widening of the main negative peak in the CAP is an indication that the increase in latency (decreased conduction velocity) affected different nerve fibers of the nerve differently. That there is only a small decrease of the amplitude of the negative peak indicates that almost all of the fibers of the auditory nerve were conducting nerve impulses. Changes in neural conduction that cause increases in the latency of the main negative peak with little change in amplitude indicate that the only effect of the surgical manipulation was an increase in neural conduction time (decrease in conduction velocity). Changes that consist of broadening of the negative peak indicate that the latency of neural conduction has increased (decreased conduction velocity) unevenly for different nerve fibers (Fig. 7.14). We believe that this is what happens when the auditory nerve is stretched moderately. Provided that proper action is taken promptly to reverse the injury, such changes seem to be completely, or nearly completely, reversible so that the patient does not acquire hearing deficits when assessed by traditional measurements of hearing postoperatively.

Recordings from the Cochlear Nucleus

Less experience has been gained regarding the interpretations of recordings made from the vicinity of the cochlear nucleus than those made from recording the auditory nerve. It is not known for certain which of the different components of the potentials that are recorded from the cochlear nucleus are most sensitive to changes in neural conduction in the auditory nerve.

The initial, fast component that signals the arrival of nerve activity at the cochlear nucleus may, therefore, be regarded to provide the same information about injury of the auditory nerve as CAP recorded from the exposed auditory vestibular nerve. The amplitudes of the slow components, however, decrease at a different rate than the initial, fast component when stimulus intensity is decreased (see page 140), which may mean that fast components are more sensitive to changes in neural conduction in the auditory nerve than are the slow components. Results from intraoperative recording during the removal of a vestibular schwannoma, such as those illustrated in Fig. 7.15, seem to support this hypothesis. The latencies of both the fast and slow components of these potentials, however, were prolonged as a result of surgical manipulation. This indicates that the latencies (but perhaps not the amplitudes) of either slow or fast components may be valid indicators of changes in neural conduction in the auditory nerve.

Effect of Injury to the Auditory Nerve on the ABR

It has traditionally been the latency of the different components of the ABR that has been used as a criterion for altered neural conduction in the auditory nerve. As discussed above, the amplitude of the CAP that can be recorded from a nerve is proportional to the number of nerve fibers that are conducting nerve impulses, and a loss of conduction in some nerve fibers causes a decrease in the amplitude of the recorded CAP. Presumably, this means that the amplitudes of the different components of the ABR also change when neural conduction in the auditory nerve is altered. It would, therefore be expected that monitoring amplitudes of the different components of the ABR in addition to monitoring latencies would be of value, and studies have supported this assumption (18).

One of the reasons why latency changes have been favored over amplitude changes as indicators of injury to the auditory nerve is that the latencies of ABR peaks are less variable than the amplitudes of the different peaks of the ABR. The reason for the greater variability of the amplitudes of the different peaks is not known, but changes in recording conditions may contribute to this variability. The noise that is always superimposed on ABR recordings also contributes to the variability of the amplitudes of the components (peaks) of the ABR.

One reason for a decrease in the amplitude of the recorded ABR is that the amplitude of the recorded potentials does actually decrease, but this is not the only reason. Another reason for a decrease in amplitude is associated with the use of signal averaging. When many responses are added, the amplitude of the resulting averaged recording decreases if the latencies of the different components (peaks) of the ABR change during the time that the responses are being collected, and the averaged response, therefore, becomes less than what it would have been if all the responses included in the average were identical.

Change in the latency of the responses that compose ABR during the time in which the evoked potentials are being collected also causes changes in the waveform of the averaged response, and the waveform of the averaged response is different from that of any waveform of the individual responses that were added to make up the averaged response. These effects of the averaging process increase when more responses are added, thus taking more time to complete the averaged response. This problem worsens as the number of changes that occur in the responses increase during the time of data acquisition. However, studies have suggested that monitoring the amplitudes of the peaks of the ABR during operations where the auditory nerve is being manipulated is valuable in detecting changes in the function of the auditory nerve (18).

Relationship Between Changes in ABR and in CAP from the Auditory Nerve and the Cochlear Nucleus

The CAP recorded from the exposed CN VIII have specific relationships to the waveform of the ABR as discussed above (page 141). Surgical manipulations of the auditory nerve that cause changes in the waveform of the CAP recorded from the exposed CN VIII also cause changes in the ABR, but the changes in the ABR are less specific and, therefore, more difficult to interpret (Fig. 7.16). Examination of the CAP recorded from the exposed CN VIII shows an increase in latency and widening of the negative peak of the CAP after surgical manipulation of CN VIII. These changes indicate that the increase in neural conduction time is different for different auditory nerve fibers, but similar information cannot be obtained from inspection of the ABR.

Surgically induced injuries to the auditory nerve result in an increase in the latency of the later (slow) components recorded from the cochlear nucleus. This change in latency is not necessarily seen in the CAP recorded from CN VIII. The same can be said for the later components of the ABR (peaks III and V). The amplitudes of these different components of auditory evoked potentials do not necessarily change to the same degree as a result of injury to the auditory nerve as do the CAP amplitudes.

One reason that the different components of the far-field response (ABR) may change in a different way than the near-field response (CAP from the auditory nerve or cochlear nucleus) is that the different components of the ABR are less dependent on the temporal coherence of neural activity than are the responses that are recorded directly from the auditory nerve. This is especially the case for the later peaks in the ABR (peaks III and V), which seem to be less dependent on temporal coherence of neural activity than the earlier peaks (peaks I and II). Thus, a large reduction in the temporal coherence of neural activity in auditory nerve fibers, which manifests as a large reduction in the response from the auditory nerve, may cause much less reduction in the amplitude of the later peaks in the ABR. This is also why the amplitude of the CAP recorded from the auditory nerve are probably more sensitive to surgically induced injuries than later peaks of the ABR such as peak V. While the CAP recorded from the auditory nerve usually have much lower amplitudes in patients with hearing loss caused by auditory nerve injuries; the amplitude of wave V in patients with this kind of hearing loss may be closer to that of patients with normal auditory nerve function. Thus, the later peaks of ABR, particularly peak V, are often less affected by injuries to the auditory nerve than earlier peaks. This means that also the CAP recorded directly from the auditory nerve are likely to be more sensitive to injury of the auditory nerve than is peak V of the ABR. This is the reason that the CAP recorded directly from CN VIII (and the initial components of the response from the cochlear nucleus) may be better indicators of injury of the auditory nerve and thus, more suitable for use in intraoperative monitoring during operations in which the eighth nerve is manipulated than recordings of the ABR when changes in peak V are used.

It was mentioned in Chap. 5 that excitation of the hair cells in the basal portion of the cochlea evokes more synchronized discharges than does excitation of hair cells that are located in the low-frequency (apical) portion of the basilar membrane. Excitation of low-frequency hair cells, therefore, contributes little to the CAP and ABR elicited by wideband click sounds. In a similar way, it may be assumed that the loss of low-frequency nerve fibers may not affect the responses to wideband click sounds noticeably, and it is possible that low-frequency hearing loss may escape detection by intraoperative monitoring when click sounds are used as stimuli.

Other Causes of Injury to the Auditory Nerve

It was discussed above how retraction and heating of the auditory nerve are probably the most common causes of injury to the auditory nerve during surgical operations in the CPA. However, there are other causes of injury that can occur during an operation. One type results from irrigating the surgical area, but there are also unknown causes of injury to the auditory nerve.

Injury to the Auditory Nerve from Irrigating.

Results of intraoperative monitoring of ABR have shown evidence that irrigation of the CPA in the region of CN VIII can cause severe injury to the auditory nerve and possibly lead to permanent hearing impairment and even deafness. It was first believed that a strong stream of fluid from a syringe used for irrigation could injure the auditory nerve, but later it was found that even a low velocity flooding of saline into the CPA could injure the auditory nerve. These experiences changed the way irrigation in the CPA was performed, and saline was gently poured on the cerebellum and never directly into the CPA.

These are examples of how intraoperative neurophysiological monitoring can improve operative techniques.

Unknown Causes of Injury to the Auditory Nerve.

Experience from intraoperative monitoring of auditory evoked potentials in MVD operations of cranial nerves has shown that there may be causes for injury to the auditory nerve other than direct and known surgical manipulations or heating from electrocoagulation.

An example of one such unknown cause of injury was a case where a patient lost hearing after an operation in the CPA during which there were no remarkable changes in the auditory evoked potentials. ABR were not monitored in the operating room after the dura was closed because it was believed then that the risk of injury to the auditory nerve had passed when the dura was closed. However, the ABR in this patient were recorded automatically to the end of the operation as a part of a research project. After it was discovered that the patient had suffered a total hearing loss, examination of the records revealed a steadily increasing latency of peak V of the ABR after the dura was closed (Fig. 7.17A). Obviously, something happened after closing the dura that caused the auditory nerve to be stretched or affected in some other way. This experience taught us to always monitor ABR until skin closure. On several occasions after this experience, once the dura was closed, large changes in the ABR occurred in similar operations. In each of these patients, reopening the dura and releasing fluid and irrigating the CPA caused the ABR to recover and thus, seemingly resolved the problem. However, it was not possible to pinpoint the exact cause of these ABR changes. None of these patients suffered permanent hearing impairment.

In similar operations in which changes in the ABR during the operation were due to operative difficulties, the latency of peak V of the ABR typically decreased toward normal values during the wound closure. Fig. 7.17B shows results from a patient who experienced large changes in evoked potentials during the operation, but the latency of peak V decreased during wound closure. The patient had a moderate postoperative hearing impairment, but the hearing improved within a 3-month period.

Practical Aspects Regarding Monitoring Auditory Evoked Potentials in Operations for Vestibular Schwannoma

Most of the examples of results of intraoperative monitoring of auditory evoked potentials that were given earlier in this chapter were from monitoring of patients who underwent MVD of cranial nerves to relieve TGN, HFS, DPV, or tinnitus. It was shown that intraoperative monitoring of auditory evoked potentials could decrease the risk of postoperative hearing loss in such patients (3). MVD operations are rare, but similar methods to preserve hearing can be used in other operations in the CPA, such as those to remove vestibular schwannoma. Such operations are much more common than MVD operations. Diagnostic methods for identifying vestibular schwannoma continue to improve, and such tumors can now be identified while still small. Many surgeons recommend surgery for small vestibular schwannoma in patients that have usable hearing to help them retain the greatest degree of this sensory function. For that, intraoperative monitoring of the function of the auditory nerve is essential.

Four different ways of monitoring auditory function in surgical operations to remove vestibular schwannoma have been described: recording of far-field auditory evoked potentials (ABR), recording of near-field auditory elicited potentials from the ear ( ECoG), CAP from the auditory nerve, and CAP from the surface of the cochlear nucleus.

Recording of ABR

An example of ABR recorded during an operation to remove a vestibular schwannoma in a patient who had good hearing before the operation (96% speech discrimination) is shown in Fig. 7.13. Despite variations in the ABR during the operation – there was an almost 1-ms prolongation of the latency of peak III in the early phase of the tumor resection procedure – the ABR obtained at the time of closure were remarkably similar to those obtained preoperatively (Fig. 7.18). Postoperatively, the patient’s speech discrimination score was 96% (recorded speech material), and his pure tone audiogram showed no significant hearing loss (except at 4 and 8 kHz) as a result of the operation.

If peak I of the ABR changes or disappears during an operation and there also is a change in all other peaks (or total obliteration of the ABR), it is a sign that the blood supply to the ear (cochlea) has been compromised. If peak I is largely unchanged while there are changes in peaks III or V, it is likely that there has been injury to the intracranial portion of the auditory nerve with the blood supply to the cochlea remaining intact. (In some patients, no peak III can be identified, as is the case in the patient whose recordings are seen in Fig. 7.18) If there is a change in peak V, but peak III is unchanged, there is reason to assume that the brainstem has been affected by surgical manipulations, or that there is ischemia due to impaired blood supply to the brainstem. If it is not possible to clearly identify peak I, a judgment about the cause of a change in, for instance, peak V of the ABR, cannot be made with certainty, and the anatomical location of the injury is less obvious.

Patients who undergo operations to remove vestibular schwannoma often have abnormal ABR before the operation (as the one seen in Fig. 7.18) because the tumor affects the neural conduction in the auditory nerve, and the components of the ABR other than peak I are delayed and often have much smaller amplitudes than normal. This results in the need to average more responses in order to obtain an interpretable recording and consequently makes it more difficult to use ABR to detect injury to the auditory nerve.

Patients undergoing operations to remove vestibular schwannoma are usually not paralyzed during the operation because the administration of muscle relaxants prevent monitoring of the facial nerve, which is critical to preserving facial nerve function. The small EMG activity of the head muscles that may occur spontaneously, or when the facial nerve is manipulated, acts as noise that contaminates the ABR recordings. This impairs the SNR of the recorded ABR and thus, increases the time required to obtain an interpretable record. There is, therefore, a great need to optimize the way ABR are recorded and processed, such as utilizing optimal stimulus and recording parameters, aggressive filtering, and an efficient quality control system that does not require any additional time for data collection (Chap. 18). By taking these matters into proper consideration, it is possible to obtain interpretable ABR and detect changes in ABR by recording for about 1–3 min, at least in patients with reasonably good ABR.

Recording from the Vicinity of the Ear

Some investigators have monitored auditory evoked potentials recorded from the ear in operations to remove vestibular schwannoma (15, 42). For direct recording from the cochlear capsule, a recording electrode must be passed through the tympanic membrane, which is an invasive procedure that takes considerable skill to perform safely. An electrode placed on the cochlear capsule not only records CAP from the distal portion of the auditory nerve, but it also records the cochlear microphonics (CM) potential and the summating potential (SP). These three different kinds of auditory evoked potentials are known as the electrocochleographic potentials (ECoG) (Fig. 7.19). Only one of the components of the ECoG is of interest in intraoperative monitoring for vestibular schwannoma, namely, the CAP from the auditory nerve.

The CAP from the auditory nerve that is recorded from the cochlear capsule usually has amplitudes within the range of several microvolts (43) and can, therefore, be evaluated with very little signal averaging (Fig. 7.14A). This makes it possible to detect changes in CAP with practically no delays.

ECoG potentials can also be recorded from a (wick) electrode placed on the tympanic membrane (Fig. 7.19B) (43) or from an electrode placed in the ear canal. However, the amplitude of the CAP component recorded this way is much smaller than those recorded from the cochlear capsule, and a considerable number of responses must be averaged before an interpretable record can be obtained.

There are, unfortunately, several problems associated with the use of ECoG potentials recorded from the ear, or its vicinity, for intraoperative monitoring of hearing in patients undergoing vestibular schwannoma surgery. These problems are related to the fact that the CAP recorded from the ear originate from the very distal portion of the auditory nerve (where it exits the cochlea), and, therefore, the ECoG potentials do not show change when the intracranial portion of the auditory nerve has actually been injured. In fact, the intracranial portion of the eighth nerve can be severed without any noticeable change occurring in the CAP recorded from the ear. Because it is the intracranial portion of the auditory nerve that is most likely to be injured during the removal of vestibular schwannoma, recordings of ECoG potentials are, therefore, not suitable for monitoring purposes in connection with operations for vestibular schwannoma because they do not detect injuries to the intracranial portion of the auditory nerve. Recording ECoG potentials makes it possible, however, to detect if the blood supply to the cochlea has been compromised. Blood supply compromise can also be detected by methods that are useful in monitoring nerve conduction in the intracranial portion of the auditory nerve such as recording from the intracranial portion of CN VIII, the cochlear nucleus, or the ABR where peak I is an indicator of the function of the ear (Fig. 7.18).

Recording CAP Directly from the Exposed Eighth Cranial Nerve

The ABR of patients with vestibular schwannoma often have small amplitudes, which, consequently, makes obtaining an interpretable record a time-consuming activity. This makes it important to be able to record CAP from the auditory nerve or the response from the cochlear nucleus because both have large amplitudes and are not easily contaminated by EMG activity. The CAP can be observed after only a few responses.

It is relatively easy to place a recording electrode on the proximal portion of the eighth nerve in operations on small vestibular schwannoma when there is a segment of the eighth nerve near the brainstem that is free of tumor (4–7). Click-evoked CAP from the eighth nerve can provide a prompt indication of injury to the auditory nerve thereby aiding in the preservation of hearing. The situation is even more apparent in operations on larger tumors where the tumor has reached the brainstem. In such operations, it is not possible to place an electrode on the proximal portion of the eighth nerve, at least not until some of the tumor has first been removed (because the eighth nerve in such cases is embedded in the tumor or is underneath it).

Recording from the Vicinity of the Cochlear Nucleus

Recording from the vicinity of the cochlear nucleus can to a great extent solve these practical problems as described above (page 135). An electrode can be placed in the lateral recess of the fourth ventricle even when operating on large vestibular schwannoma (see Fig. 7.10). More importantly, an electrode placed in or near the foramen of Luschka is far away from the operative field, and the electrode is not as easily dislodged as may be the case when it is placed on CN VIII. This technique is an effective way of monitoring neural conduction in the auditory nerve.

Effect of Drilling of Bone

There are three ways that drilling of the bone for exposing the content of the internal auditory meatus in operations such as those for vestibular schwannoma can affect the ABR or the potentials recorded from the auditory nerve or the cochlear nucleus.

1.
The response may decrease or even disappear totally because the bone-conducted noise masks the sounds used to elicit the auditory evoked potentials. This noise is transmitted to the cochlea through vibrations in the skull bone (bone conduction) rather than via the normal route for airborne sound (through the middle ear). Although sealing the ear canal reduces the airborne noise that reaches the tympanic membrane, it does not reduce the noise from drilling that reaches the cochlea through bone conduction. In fact, a closed ear canal may enhance the transmission of bone-conducted sound to the cochlea, although this effect is slight. In any event, the stimulation of the cochlea from the noise produced by drilling is usually so strong that it is impossible to record auditory evoked potentials during drilling.
2.
Intensive drilling of the internal auditory meatus may cause impairment of the function of the cochlea that may in turn cause a temporary (or permanent) threshold shift because the drilling noise overloads the cochlea. This may cause alterations in the ABR, and other auditory evoked potentials may be affected for some time after termination of the drilling or permanently.
3.
The drilling may heat the bone, and that heat may be transmitted to the auditory nerve in the internal auditory meatus where it may cause temporary or permanent injury to the auditory nerve. This damage is revealed in changes in ABR, CAP recorded from the cochlear nucleus, as well as in the CAP recorded from the exposed auditory nerve. It is, therefore, important to monitor auditory evoked potentials during pauses in the drilling.

Factors Other than Surgical Manipulation that May Influence Auditory Evoked Potentials

Monitoring of ABR and CAP from CN VIII or the cochlear nucleus is affected by the conditions of the ear and the auditory nervous system of individual patients before the operation. Prior operations affecting the same structures, such as CN VIII, and the patient’s general health condition can affect the recorded evoked potentials. The presence of other disorders, such as cardiovascular disorders, can also affect auditory evoked potentials.

Effects of Preoperative Hearing Loss on ABR and CAP from the Auditory Nerve

The presence of preoperative hearing loss may affect click-evoked ABR as well as the CAP that can be recorded from the exposed CN VIII or the vicinity of the cochlear nucleus. The effect depends on the degree and type of hearing loss. Hearing loss that is caused by an impairment of conduction of sound to the cochlea (affecting the ear canal, ear drum, middle ear) (30) affects the ABR and CAP from the auditory nerve and the cochlear nucleus in a similar way as does a decrease in the intensity of the stimulus sound. Different forms of conductive hearing loss may affect sound transmission for different frequencies differently and may, thereby affect the recorded responses dissimilarly. Evoked responses from the auditory nervous system to broad spectrum sounds, such as click sounds, may, therefore, differ between a person with hearing loss and a person with normal hearing, even when the stimulus intensity has been elevated to compensate for the loss in sound transmission to the cochlea.

The high-frequency spectral components of broadband sounds (such as click sounds) are most important for eliciting auditory evoked responses. Low-frequency hearing loss of the conductive type may, therefore, not affect the ABR noticeably, and individuals with such hearing loss may have ABR that are similar to those of individuals with normal hearing. The intensity of the click sound that is used to elicit ABR intraoperatively in a patient with conductive hearing loss should, therefore, only be increased if the hearing loss includes the high-frequency range of hearing (above 4 kHz). If a true conductive hearing loss involves the high-frequency range of hearing, the stimulus sound level can be increased by an amount equal to the conductive hearing loss for high frequencies (4–8 kHz) in order to obtain an interpretable ABR recording. It is, however, unusual that conductive hearing loss extends to the high-frequency range of hearing.

A moderate sensorineural hearing loss caused by cochlear deficits has minimal effects on the ABR. Sensorineural hearing loss often occurs in elderly individuals (presbycusis), but may also be present in younger individuals and is often caused by noise exposure (NIHL) or administration of ototoxic drugs, such as aminoglycoside antibiotics. These factors all affect auditory sensitivity to sounds of higher frequencies more than they affect sounds of lower frequencies. Cochlear hearing loss is caused by the loss of outer hair cells, primarily, in the basal portion of the cochlea and thus, mostly affecting high-frequency hearing. More important perhaps is the fact that the loss of outer hair cells affects the cochlear amplifier, which is most important for sounds of low intensity, but the loss usually does not affect cochlear function noticeably at the high sound levels used for recording auditory evoked potentials (30). While hearing loss of cochlear origin can affect the waveform of ABR, there is no reason to increase the stimulus intensity used to elicit auditory evoked potentials in patients who have a cochlear type of hearing loss.

Such hearing loss may also affect the CAP recorded from the exposed CN VIII to an extent that depends on the severity of the hearing loss (see Fig. 7.9). The CAP that are recorded from patients with such hearing loss often has a more complex waveform than in individuals with normal hearing with several peaks (44, 45).

In the extreme situation in which a disorder of the ear or of the auditory nervous system is so severe that it is not possible to obtain an interpretable ABR recording from the patient before the operation; it is not possible to perform intraoperative monitoring of auditory evoked potentials. If the person in charge of monitoring did not know before the operation that such a patient had a severe hearing loss, a tedious search for technical causes for the failure to obtain reproducible ABR in the operating room would ensue. On the other hand, if the patient had reproducible ABR preoperatively, but it is not possible to obtain a response in the operating room, then it is obvious that the cause of the failure to obtain reproducible ABR in the operating room is a technical problem that must be solved before the operation can begin.

Previous Injuries to CN VIII

The ABR recorded from people with hearing losses caused by injury to the auditory nerve may have complex abnormalities, including increased interpeak latencies, and the waveforms of the recorded potentials are different from those that are seen in patients with normal hearing. Injury to the auditory nerve is typically present before the operation in patients with vestibular schwannoma and in patients who have undergone previous surgical operations in which injury to the auditory nerve has occurred. Such conditions affect the ABR in a different way than do lesions to the cochlea. Injuries to the auditory nerve typically result in ABR with low amplitudes and complex waveforms. The CAP recorded from the exposed CN VIII in patients with an injured auditory nerve is likely to have complex waveforms (Fig. 7.9).

Slight injury to the auditory nerve may decrease the temporal coherence of discharges in different nerve fibers because the conduction velocity in different fibers may be affected differently as a result of such injury. The complex waveform and low amplitude of the CAP in patients with an injured auditory nerve is a result of a difference in the conduction velocity of the nerve fibers that make up the auditory nerve.

Relationship Between Auditory Evoked Potentials and Hearing Acuity

It is important to remember that changes in auditory evoked potentials do not measure changes in hearing. The effects on hearing thresholds from injuries to the auditory nerve can, therefore, not be predicted directly on the basis of knowledge about the changes in the CAP recorded from the auditory nerve in response to loud click sounds, as is commonly the case for eliciting the response in the operating room. The ability to understand speech, which is more important than the hearing threshold, is even more difficult to predict on the basis of evoked potentials. While there is a correlation between hearing threshold (audiogram) and speech discrimination scores when the hearing loss is caused by damage to the sensory cells in the cochlea, such a relationship does not exist for damage to the auditory nerve.

Individuals in whom the intracranial portion of the auditory nerve has sustained surgically induced injury often have severely impaired speech discrimination, with only a moderate reduction in hearing threshold, as revealed by pure tone audiograms (Fig. 7.20). This is probably because injuries to the auditory nerve impair the timing of auditory nerve activity and synchronization of neural activity in the auditory nerve is important for speech discriminations, but has less effect on pure tone thresholds reflected in the audiogram. Individuals who have suffered injuries to their auditory nerve often have severe tinnitus, which causes a severe reduction of the quality of life. After injury to the auditory nerve, it is not known if deterioration of the earliest peaks of the ABR with a preservation of peak V means that the patient’s ability to understand speech will be impaired, or if also peak V must be noticeably affected before a functional change in hearing may occur.

Even though changes in neural conduction that occur during manipulation of the auditory nerve (as revealed by changes in the CAP recorded from the exposed auditory nerve) may have been totally reversed, the auditory nerve may still have suffered some injury. Thus, studies in animals indicate that the injury that is caused by a partial dislocation of the transition zone between the peripheral and central myelin of the auditory nerve (Obersteiner–Redlich zone, or O–R zone) (46–48) may be reversible, but still may imply permanent injury. This may impair speech discrimination without causing any noticeable abnormality in auditory evoked potentials and without noticeable change in hearing, thresholds as reflected in the audiogram.

Injuries to the auditory nerve from surgical manipulations often produce a greater loss in speech discrimination than would have been inferred from the threshold elevation to pure tones, (pure tone audiograms) (Fig. 7.20 (3 30)). The likely reason is that slight injuries to the auditory nerve may cause reduced temporal coherence of neural firing in auditory nerve fibers without affecting the threshold to pure tones (thus having a normal pure tone audiogram). Deterioration of the timing of neural discharges, which may occur from injury to the auditory nerve, is known to affect the ability to discriminate speech.

The effects of injuries to the auditory nerve on everyday use of hearing (such as for speech communication) are not well described by the pure tone audiogram because injury to the auditory nerve is likely to cause a considerable decrease in the speech discrimination score even when the pure tone threshold is only slightly affected as indicated by the conventional audiogram (30, 49). Since speech discrimination can deteriorate to a considerable degree with little or only moderate changes of the pure tone audiogram (30, 48), the pure tone audiogram alone is not a suitable measure of (functional) hearing loss in patients whose CN VIII has been injured; speech discrimination tests should instead be used to evaluate injuries to the auditory nerve (49).

Other Advantages of Recording Auditory Evoked Potentials Intraoperatively

Studies of the changes in auditory evoked potentials have provided information that have resulted in the development of better surgical methods, which are important for the surgeon performing the surgical procedure and for the individual patient in whom monitoring was performed. Thus, there are advantages from recording of CAP directly from the auditory nerve other than for reducing the risk of hearing loss in the individual patient. Recordings of CAP directly from the exposed eighth nerve or the vicinity of the cochlear nucleus during operations in the CPA have not only been valuable in reducing injuries due to surgical manipulations in individual patients, but they have also contributed to our understanding of how injuries to nerves from surgical manipulations may come about. The ability to detect changes in neural conduction, almost instantaneously, has made it possible to detect such changes early enough to be able to identify exactly which step in an operation caused an adverse effect on neural conduction. For example, this technique has made it possible to relate the effects to specific surgical events, such as electrocoagulation to injury of the auditory nerve, because recording of the CAP from the auditory nerve or the cochlear nucleus has made it possible to determine exactly which step in an operation caused a change in function of the auditory nerve. Such observation showed that the risk that a nerve can be damaged by the heat used in electrocoagulation of blood vessels is greater than earlier believed. Such studies would not have been possible using recordings of ABR because of the time it takes to obtain an interpretable record.

Experience has demonstrated that the auditory nerve can be seriously injured by the normal use of bipolar electrocoagulation when performed close to the auditory portion of CN VIII. The adverse effect on the auditory nerve is not caused by a spread of high-frequency current (which was a serious problem when monopolar coagulation was used), but rather by the spread of heat. Since all electrocoagulation is based on heating the tissue in question (usually a vein), such heat may spread to neural tissue located close to the site that is undergoing coagulation. Electrocoagulation using the bipolar technique may, thus, injure neural tissue from the spread of heat used to coagulate nearby tissue, even though the spread of high-frequency current may be negligible.

These findings have prompted a change in the way electrocoagulation is performed near the eighth nerve, namely, to use the lowest possible current, to do electrocoagulation in bursts of only a few seconds duration and to allow time for the tissue to cool between periods of electrocoagulation. These changes in the way blood vessels are coagulated have reduced the risks of injury to neural tissue from electrocoagulation in general.

Recordings of CAP have also provided information on how the auditory nerve may be injured by stretching, and how the nerve is highly sensitive to heat (from electrocoagulation).

Recording of the CAP from the auditory nerve has also shown that there are considerable differences in individual susceptibility to mechanical manipulation of the auditory nerve. In operations in the CPA using the retromastoid approach, manipulations of the eighth nerve may occur, for instance, when the cerebellum is retracted. It has been indicated in earlier studies that medial-to-lateral retraction (50, 51) places the eighth nerve at greater risk than does retraction in a caudal-to-rostral direction. This hypothesis has been confirmed by studies of CAP recordings from the auditory nerve (2).

Animal experiments have helped to understand how certain surgical manipulations can cause injury to the auditory nerve, and it has been shown that injuries are likely to occur where the auditory nerve passes through the cribriform plate (46, 47, 52).

Experience from intraoperative monitoring has also shown that the arachnoid membrane that covers CN VIII may be stretched by retracting the cerebellum and thereby, stretch the eighth nerve. It was found that changes in auditory evoked potentials that occur during MVD operations can be reduced by opening the arachnoid membrane widely as soon as possible after it has been exposed (Jho and Møller, unpublished observation 1990) even in operations in which only CN V must be exposed in order to carry out the operation. The reason that it is beneficial to make a large opening in the arachnoid membrane is probably that tensions along the edge of the opening are reduced. It is also possible that the arachnoidal membrane that is connected to CN VIII can stretch the auditory nerve when, for example, the cerebellum is retracted.

These are examples of how intraoperative neurophysiological monitoring can promote the development of surgical methods that are more effective and have less risk.

Recording of ABR can also be useful for other purposes than monitoring the auditory system. The use of recordings of ABR has been shown to be of value for detecting brainstem manipulations that affect control of blood pressure and heart rate (Chap. 11, Figs. 11.9 and 11.10).

Furthermore, recordings of auditory evoked potentials have provided important basic information about the function of the normal auditory system and about some pathological conditions such as tinnitus.

Anesthesia Requirements

Although slight changes in ABR have been reported as a result of the administration of certain anesthetic agents (53, 54), the ABR are remarkably insensitive to anesthesia. The type of anesthesia can be chosen without any consideration as to whether or not ABR are to be monitored. However, it has been noted that the patient’s body temperature has a significant effect on the latency of ABR. When the body temperature drops below 35.0°C, there is a noticeable increase in the latency of the peaks of the ABR (55). This should be remembered when interpreting slow changes in ABR.

Notes

1.
Radio Shack Corporation, Ft. Worth, Texas 76102.
2.
3M™ Blenderm™ Surgical Tape, 3M Global Headquarters 3M Corporate Headquarters 3M Center) St. Paul, MN 55144-1000.
3.
Medwire Corporation, 121 South Columbus Avenue, Mt. Vernon, New York 10553.

References

Grundy B (1983) Intraoperative monitoring of sensory evoked potentials. Anesthesiology 58:72–87.
Article CAS PubMed Google Scholar
Møller AR and PJ Jannetta (1983) Monitoring auditory functions during cranial nerve microvascular decompression operations by direct recording from the eighth nerve. J. Neurosurg. 59:493–9.
Article PubMed Google Scholar
Møller AR and MB Møller (1989) Does intraoperative monitoring of auditory evoked potentials reduce incidence of hearing loss as a complication of microvascular decompression of cranial nerves? Neurosurgery 24:257–63.
Article PubMed Google Scholar
Colletti V, A Bricolo, FG Fiorino et al (1994) Changes in directly recorded cochlear nerve compound action potentials during acoustic tumor surgery. Skull Base Surg. 4:1–9.
Article CAS PubMed Google Scholar
Silverstein H, H Norrell and S Hyman (1984) Simultaneous use of CO2 laser with continuous monitoring of eighth cranial nerve action potential during acoustic neuroma surgery. Otolaryngol. Head Neck Surg. 92:80–4.
CAS PubMed Google Scholar
Fischer C (1989) Brainstem auditory evoked potential (BAEP) monitoring in posterior fossa surgery, in Neuromonitoring in Surgery, J Desmedt Editor. Elsevier Science Publishers: Amsterdam. 191–218.
Google Scholar
Linden R, C Tator, C Benedict et al (1988) Electro-physiological monitoring during acoutic neuroma and other posterior fossa surgery. J. Sci. Neurol. 15:73–81.
CAS Google Scholar
Kuroki A and AR Møller (1995) Microsurgical anatomy around the foramen of Luschka with reference to intraoperative recording of auditory evoked potentials from the cochlear nuclei. J. Neurosurg. 82:933–9.
Article CAS PubMed Google Scholar
Møller AR, HD Jho and PJ Jannetta (1994) Preservation of hearing in operations on acoustic tumors: An alternative to recording BAEP. Neurosurgery 34:688–93.
Article PubMed Google Scholar
Raudzens PA (1982) Intraoperative monitoring of evoked potentials. Ann. N.Y. Acad. Sci. 388:308–26.
Article CAS PubMed Google Scholar
Møller AR and PJ Jannetta (1983) Auditory evoked potentials recorded from the cochlear nucleus and its vicinity in man. J. Neurosurg. 59:1013–8.
Article PubMed Google Scholar
Møller AR, PJ Jannetta and HD Jho (1994) Click-evoked responses from the cochlear nucleus: A study in human. Electroencephalogr. Clin. Neurophysiol. 92:215–24.
Article PubMed Google Scholar
Sekiya T, T Iwabuchi, T Hatayama et al (1991) Vestibular nerve injury as a complication of microvascular decompression. Neurosurgery 29:773–5.
Article CAS PubMed Google Scholar
Levine RA, WW Montgomery, RG Ojemann et al (1978) Evoked potential detection of hearing loss during acoustic neuroma surgery. Neurology 28:339.
Google Scholar
Levine RA, RG Ojemann, WW Montgomery et al (1984) Monitoring auditory evoked potentials during acoustic neuroma surgery. Ann. Otol. Rhinol. Laryngol. 93:116–23.
CAS PubMed Google Scholar
Sekhar LN and AR Møller (1986) Operative management of tumors involving the cavernous sinus. J. Neurosurg. 64:879–89.
Article CAS PubMed Google Scholar
Møller AR (1987) Electrophysiological monitoring of cranial nerves in operations in the skull base, in Tumors of the Cranial Base: Diagnosis and Treatment, LN Sekhar and VL Schramm Jr, Editors. Futura Publishing Co: Mt. Kisco, New York. 123–32.
Google Scholar
Hatayama T and AR Møller (1998) Correlation between latency and amplitude of peak V in brainstem auditory evoked potentials: Intraoperative recordings in microvascular decompression operations. Acta Neurochir. (Wien) 140:681–7.
Article CAS Google Scholar
Chiappa K, KG Gladstone and RR Young (1979) Brainstem evoked responses: Studies of waveform variations in 50 normal human subjects. Arch. Neurol. 36:81–6.
CAS PubMed Google Scholar
Campbell KCM and PJ Abbas (1987) The effect of stimulus repetition rate on the auditory brainstem response in tumor and non-tumor patients. J. Speech Hear. Res. 30:494–502.
CAS PubMed Google Scholar
Pratt H and H Sohmer (1976) Intensity and rate functions of cochlear and brainstem evoked responses to click stimuli in man. Arch. Otol. Rhinol. Laryngol. 212:85–92.
Article CAS Google Scholar
Møller AR (1986) Effect of click spectrum and polarity on round window N₁N₂ response in the rat. Audiology 25:29–43.
Article PubMed Google Scholar
Scherg M and D von Cramon (1985) A new interpretation of the generators of BAEP waves I V: Results of a spatio temporal dipole. Electroencephalogr. Clin. Neurophysiol. 62:290–9.
Article CAS PubMed Google Scholar
Jewett DL (1987) The 3 channel Lissajous’ trajectory of the auditory brain stem response. IX. Theoretical aspects. Electroencephalogr. Clin. Neurophysiol. 68:386–408.
Article Google Scholar
Gardi JN, YS Sininger, WH Martin et al (1987) The 3-channel Lissajous’ trajectory of the auditory brain-stem response. IV. Effect of lesions in the cat. Electroencephalogr. Clin. Neurophysiol. 68:360–7.
Article Google Scholar
Pratt H, Z Har’el and E Golos (1984) Geometrical analysis of human three-channel Lissajous’ trajectory of auditory brain stem evoked potentials. Electroencephalogr. Clin. Neurophysiol. 58:83–8.
Article CAS PubMed Google Scholar
Pratt H, N Bleich and WH Martin (1985) Three channel Lissajous’ trajectory of humans auditory brain stem evoked potentials. I. Normative measures. Electroencephalogr. Clin. Neurophysiol. 61:530–8.
Article CAS PubMed Google Scholar
Møller AR (1988) Evoked Potentials in Intraoperative Monitoring. Williams and Wilkins: Baltimore.
Google Scholar
Møller AR (1988) Use of zero-phase digital filters to enhance brainstem auditory evoked potentials (BAEPs). Electroencephalogr. Clin. Neurophysiol. 71:226–32.
Article PubMed Google Scholar
Møller AR (2006) Hearing: Anatomy, Physiology, and Disorders of the Auditory System, 2nd Ed., Academic Press: Amsterdam.
Google Scholar
Møller AR and PJ Jannetta (1981) Compound action potentials recorded intracranially from the auditory nerve in man. Exp. Neurol. 74:862–74.
Article PubMed Google Scholar
Hashimoto I, Y Ishiyama, T Yoshimoto et al (1981) Brainstem auditory evoked potentials recorded directly from human brain stem and thalamus. Brain 104:841–59.
Article CAS PubMed Google Scholar
Hashimoto I (1982) Auditory evoked potentials recorded directly from the human VIIIth nerve and brain stem: Origins of their fast and slow components. Electroencephalogr. Clin. Neurophysiol. Suppl. 36:305–14.
CAS PubMed Google Scholar
Hashimoto I (1982) Auditory evoked potentials from the humans midbrain: Slow brain stem responses. Electroencephalogr. Clin. Neurophysiol. 53:652–7.
Article CAS PubMed Google Scholar
Møller AR, PJ Jannetta and MB Møller (1981) Neural generators of brainstem evoked potentials. Results from human intracranial recordings. Ann. Otol. Rhinol. Laryngol. 90:591–6.
PubMed Google Scholar
Møller AR and PJ Jannetta (1982) Evoked potentials from the inferior colliculus in man. Electroencephalogr. Clin. Neurophysiol. 53:612–20.
Article PubMed Google Scholar
Spire JP, GJ Dohrmann and PS Prieto (1982) Correlation Of Brainstem Evoked Response with Direct Acoustic Nerve Potential, J Courjon, F Manguiere and M Reval, Editors. Vol. 32. Raven Press: New York.
Google Scholar
Møller AR (1993) Direct eighth nerve compound action potential measurements during cerebellopontine angle surgery, in Proceedings of the First International Conference on ECoG, OAE, and Intraoperative Monitoring, D Höhmann, Editor. Kugler Publication: Amsterdam, The Netherlands. 275–80.
Google Scholar
Møller AR and PJ Jannetta (1982) Auditory evoked potentials recorded intracranially from the brainstem in man. Exp. Neurol. 78:144–57.
Article PubMed Google Scholar
Terr LI and BJ Edgerton (1985) Surface topography of the cochlear nuclei in humans: Two and three-dimensional analysis. Hear. Res. 17:51–9.
Article CAS PubMed Google Scholar
Hatayama T, T Sekiya, S Suzuki et al (1999) Effect of compression on the cochlear nerve: A short-and long-term electrophysiological and histological study. Neurol. Res. 21:591–610.
Google Scholar
Ojemann RG, RA Levine, WM Montgomery et al (1984) Use of intraoperative auditory evoked potentials to preserve hearing in unilateral acoustic neuroma removal. J. Neurosurg. 61:938–48.
Article CAS PubMed Google Scholar
Winzenburg SM, RH Margolis, SC Levine et al (1993) Tympanic and transtympanic electrocochleography in acoustic neuroma and vestibular nerve section surgery. Am. J. Otol. 14:63–9.
CAS PubMed Google Scholar
Møller AR and HD Jho (1991) Effect of high frequency hearing loss on compound action potentials recorded from the intracranial portion of the human eighth nerve. Hear. Res. 55:9–23.
Article PubMed Google Scholar
Møller AR and HD Jho (1991) Compound action potentials recorded from the intracranial portion of the auditory nerve in man: Effects of stimulus intensity and polarity. Audiology 30:142–63.
Article PubMed Google Scholar
Sekiya T and AR Møller (1987) Avulsion rupture of the internal auditory artery during operations in the cerebellopontine angle: A study in monkeys. Neurosurgery 21:631–7.
Article CAS PubMed Google Scholar
Sekiya T and AR Møller (1987) Cochlear nerve injuries caused by cerebellopontine angle manipulations. An electrophysiological and morphological study in dogs. J. Neurosurg. 67:244–9.
Article CAS PubMed Google Scholar
Møller AR (2006) Neural Plasticity and Disorders of the Nervous System. Cambridge: Cambridge University Press
Book Google Scholar
Møller AR (1993) Late waves in the response recorded from the intracranial portion of the auditory nerve in humans. Ann. Otol. Rhinol. Laryngol. 102:945–53.
PubMed Google Scholar
Gardner WJ (1962) Concerning the mechanism of trigeminal neuralgia and hemifacial spasm. J. Neurosurg. 19:947–58.
Article CAS PubMed Google Scholar
Jannetta P, (1981) Hemifacial spasm, in The Cranial Nerves, M Samii and P Jannetta, Editors. Springer: Heidelberg, West Germany. 484–93.
Google Scholar
Møller AR and T Sekiya (1988) Injuries to the auditory nerve: A study in monkeys. Electroencephalogr. Clin. Neurophysiol. 70:248–55.
Article PubMed Google Scholar
Cohen MS and RH Britt (1982) Effects of sodium pentobarbital, ketamine, halothane, and chloralose on brain stem auditory evoked responses. Anesth. Analg. 61:338–43.
Article CAS PubMed Google Scholar
Thornton C, DM Catley, C Jorden et al (1981) Enflurane increases the latency of early components of the auditory evoked response in man. Br. J. Anaesth. 53:1102–3.
Google Scholar
Stockard JJ, F Sharbrough and J Tinker (1978) Effects of hypothermia on the human brain stem auditory response. Ann. Neurol. 3:368–70.
Article CAS PubMed Google Scholar
Martin WH, H Pratt and JW Schwegler (1995) The origin of the human auditory brainstem response wave II. Electroencephalogr. Clin. Neurophysiol. 96:357–70.
Article CAS PubMed Google Scholar
Lang J (1985) Anatomy of the brainstem and the lower cranial nerves, vessels, and surrounding structures. Am. J. Otol. Suppl, Nov:1–19.
Google Scholar
Silverstein H, H Norrell, T Haberkamp et al (1986) The unrecognized rotation of the vestibular and cochlear nerves from the labyrinth to the brain stem: Its implications to surgery of the eighth cranial nerve. Otolaryngol. Head Neck Surg. 95:543–9.
CAS PubMed Google Scholar
Møller AR (1994) Monitoring techniques in cavernous sinus surgery, in Monitoring Techniques in Neurosurgery, Chap. 15, CM Loftus and VC Traynelis, Editors. McGraw-Hill, Inc: New York. 141–55.
Google Scholar

Download references

Author information

Authors and Affiliations

School of Brain and Behavioral Sciences, The University of Texas at Dallas, 800 W. Campbell Road, Richardson, TX, 75080, USA
Aage R. Møller PhD (DMedSci)

Authors

Aage R. Møller PhD (DMedSci)
View author publications
You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Aage R. Møller PhD (DMedSci) .

Rights and permissions

Reprints and permissions

Copyright information

About this chapter

Cite this chapter

Møller, A.R. (2011). Monitoring Auditory Evoked Potentials. In: Intraoperative Neurophysiological Monitoring. Springer, New York, NY. https://doi.org/10.1007/978-1-4419-7436-5_7

Download citation

DOI: https://doi.org/10.1007/978-1-4419-7436-5_7
Published: 21 October 2010
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4419-7435-8
Online ISBN: 978-1-4419-7436-5
eBook Packages: MedicineMedicine (R0)

Publish with us

Policies and ethics

Monitoring Auditory Evoked Potentials

Abstract

Keywords

Introduction

Auditory Brainstem Responses

How to Obtain an Interpretable Record in the Shortest Possible Time?

Stimulus Intensity.

Stimulus Repetition Rate.

Sound Delivery.

Electrode Placement.

Types of Electrodes.

Processing of Recorded ABR

Filtering of Recorded Potentials.

Quality Control.

Display of ABR in the Operating Room

Recording of Near-Field Potentials

Direct Recording from the Eighth Cranial Nerve

The Anatomy of CN VIII.

Recording from the Vicinity of the Cochlear Nucleus

Interpretation of Changes in Auditory Responses

Interpretation of Changes in the ABR

Interpretations of CAP from CN VIII and the Cochlear Nucleus

Recordings from the Cochlear Nucleus

Effect of Injury to the Auditory Nerve on the ABR

Relationship Between Changes in ABR and in CAP from the Auditory Nerve and the Cochlear Nucleus

Other Causes of Injury to the Auditory Nerve

Injury to the Auditory Nerve from Irrigating.

Unknown Causes of Injury to the Auditory Nerve.

Practical Aspects Regarding Monitoring Auditory Evoked Potentials in Operations for Vestibular Schwannoma

Recording of ABR

Recording from the Vicinity of the Ear

Recording CAP Directly from the Exposed Eighth Cranial Nerve

Recording from the Vicinity of the Cochlear Nucleus

Effect of Drilling of Bone

Factors Other than Surgical Manipulation that May Influence Auditory Evoked Potentials

Effects of Preoperative Hearing Loss on ABR and CAP from the Auditory Nerve

Previous Injuries to CN VIII

Relationship Between Auditory Evoked Potentials and Hearing Acuity

Other Advantages of Recording Auditory Evoked Potentials Intraoperatively

Anesthesia Requirements

Notes

References

Author information

Authors and Affiliations

Corresponding author

Rights and permissions

Copyright information

About this chapter

Cite this chapter

Download citation

Share this chapter

Publish with us

Search

Navigation