28.1 Introduction

Intraoperative neurophysiologic monitoring of the cochlear division of the eighth cranial nerve (CN) provides the microsurgeon a tool to assess the impact of operative manipulations on the physiologic integrity and long-term function of the auditory system during surgical treatment of a vestibular schwannoma (VS). There are several options available to the surgeon to monitor the entire auditory system or specific portions thereof, each with advantages and disadvantages. Currently available options include auditory brainstem response/brainstem auditory evoked response (ABR/BAER), cochlear nerve action potential (CNAP), electrocochleography (ECoG), otoacoustic emissions, and cochlear blood flow estimates.s. Literatur ^,​ s. Literatur ^,​ s. Literatur Several groups have utilized combinations of monitoring technique including BAER and ECoGs. Literatur ^,​ s. Literatur and BAER and CNAP.s. Literatur ^,​ s. Literatur Each monitoring technique has its proponents and published reports extolling its respective virtue. However, the two most commonly employed techniques in contemporary microsurgical treatment of VS are intraoperative BAER and CNAP. This chapter focuses on these two techniques.

28.2 Background

ABR and cochlear nerve compound potential testing measure two types of auditory evoked potentials (AEPs): electrical potentials generated by populations of neurons responding to acoustic stimuli. Various terms are used to describe each response. In this chapter, the terms brainstem auditory evoked response (BAER) and cochlear nerve action potential (CNAP) will be used. Both BAER and CNAP monitoring provide measures of the physiological functioning of the auditory system, which can also be used to infer some aspects regarding the health of the underlying neuroanatomy. In the setting of microsurgery, these responses can help guide the surgeon in making manipulations that are likely to allow for the preservation of residual auditory functioning.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur However, neither response is a measurement of hearing per se. Hearing is the perception and processing of the acoustic environment. Complete and permanent loss of AEP responses can still be associated with preserved hearing.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur Additionally, preservation of responses does not guarantee postoperative hearing.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur The utility of AEPs, including BAER and CNAP, comes from the objective nature of the responses and the relative insensitivity to general anesthesia and commonly utilized muscle relaxants.s. Literatur ^,​ s. Literatur Additionally, when properly performed, these responses can give an estimate of potentially reversible damage to the auditory structures at risk or prevent further injury during surgical treatment of VS.s. Literatur

Candidacy for hearing preservation microsurgery and outcomes are discussed in other sections, including Chapters 36, 37, and 53, and are not discussed in this chapter. However, the capacity to monitor hearing physiology can be a consideration in approach selection, though typically a small contributing factor overall. In properly selected and counseled patients, a hearing-preserving approach can be undertaken without intraoperative monitoring and successful hearing preservation can be a reasonable aspiration.s. Literatur Indeed, the first report of hearing preservation after removal of a VS occurred in 1954, well before the description of the BAER in the early 1970s.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur

Both BAER and CNAP require sufficient residual physiological functioning of the peripheral and central auditory system as these structures serve as the electrical generators of the potentials being monitored. The outer, middle, and inner ear structures, except in rare circumstances, are typically unaffected by a VS whose size and residual hearing allow for the consideration of a hearing preservation treatment option. If the patient has conductive, sensory, or mixed hearing losses, these losses may preclude the generation of waveforms to monitor regardless of the impact of the VS on the auditory physiology.

28.3 General AEP Recording Principles

The recording techniques used in AEP monitoring and recording, including BAER and CNAP, are well established and share common characteristics. These AEP signals are generated by synchronized firing of large populations of acoustically sensitive neurological structures, mainly outer and inner hair cells (transducer potentials) and the neurons of the central auditory system (action potentials).s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur Typically, an acoustic stimulus is presented to the outer ear and the voltage changes that result from the acoustically responsive elements are measured over a specific time epoch from the stimulus onset. Each stage of acoustic processing generates its own response that can be characterized by morphology (shape, polarity, and amplitude of the response) and timing in relationship to the stimulus (latency).s. Literatur These potentials are very small, typically in the microvolt or nanovolt range.s. Literatur ^,​ s. Literatur To reliably identify these potentials, multiple technical and processing strategies are employed. There are various electrode configurations in the published literatures. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur; the most common montages consist of the recording (AKA active or positive) electrode at the vertex (Cz in standard “10–20 system” electroencephalogram nomenclature) and the reference (AKA inactive or negative) electrode at or below the level of the CN (earlobe or ipsilateral mastoid, designated A1 or A2).s. Literatur ^,​ s. Literatur The ground electrode can be placed in various locations such as the midline forehead (Fz) or one of the deltoid regions. Fig. 28‑1 demonstrates the typical electrode montage used at the University of Iowa during intraoperative eighth nerve monitoring. Electrodes can be skin surface contacts or subcutaneous needles. Intraoperative AEP monitoring can utilize either, but subcutaneous needle electrodes typically have lower and more stable impedances in our experience. Proper recording requires low impedances at each electrode and very small difference between all three electrodes (active, inactive, and reference). High impedances and significant differences can reduce or alter the size and morphology of the AEPs.

AEPs have many responses (often called waves) that can be recorded starting with responses from the cochlea and extending to the neocortex.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur Intraoperative BAER and CNAP focus on five main waveforms: vertex positive waves I–V.s. Literatur ^,​ s. Literatur Each wave represents synchronized spiking of a population of neurons at progressively more central locations in the auditory system.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur Each waveform has a characteristic morphology and latency, and normative data are used to class responses as normal or abnormal. Additionally, the timing between successive waves (interwave latency) can be measured and compared with normative data. When using the electrode configurations demonstrated in Fig. 28‑1 , there are five vertex (Cz) positive waveforms that constitute the BAER, labeled with Roman numerals as per Jewett.s. Literatur The timescale is measured in milliseconds from the stimulus onset. Fig. 28‑2 demonstrates a preoperative BAER recording from a patient with a 1-mm VS, which was classified as normal when compared to normative data. As reviewed in Martin and Stecker, the generators of each wave in human subjects have been postulated and are generally accepted.s. Literatur Wave I represents the synchronous activation of type I afferents laterally in the modiolus with the onset of an acoustic stimulus. This has been termed the compound action potential and can be demonstrated with CNAP, BAER, and ECoG.s. Literatur Wave II is thought to represent the proximal CN and possibly some responding cochlear nucleus neurons. Wave III is thought to represent neuronal responses in the cochlear nucleus and superior olivary complex (brainstem structures). Wave IV is thought to be produced by generators in the lateral lemniscus. Wave V is the result of spiking in the inferior colliculus (midbrain structures). There is overlap in time and space of these potentials and thus each waveform likely has some contribution from multiple generator sources. The waveforms that are utilized most frequently during intraoperative eighth nerve monitoring are waves I and V.s. Literatur In general, wave I gives a good estimate of the cochlear function (including the modiolar nerve) and wave V an estimate of the brainstem auditory function.s. Literatur A thorough review of the science behind the generators is beyond the scope of this chapter.

Given the small size of the recorded potentials, AEPs are exquisitely sensitive to inference from other intrinsic and extrinsic electrical sources such as skeletal muscle activity or electric instrumentation, respectively. Several strategies are used to reduce the impact of other electrical signals: averaging, artifact rejection, and filtering.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur Averaging consists of performing multiple individual recordings (i.e., trials), using the same stimulus type and intensity, and then calculating the mean voltage at each time point in the trial. Signals that are synchronized to the start of the recording epoch (i.e., the AEP of interest) will remain; other nonsynchronized signals will be minimized or eliminated. For large and uniform AEPs, a low number of trials are needed to obtain a reliable response pattern. The higher the number of trials needed, the longer it will take to obtain the response pattern. Artifact rejection is a process where trials with large swings in electrical potential are automatically excluded from the averaging process due to the likelihood that these records contain significant interfering signals.s. Literatur Artifact rejection parameters can be adjusted to be more or less selective in which records are excluded. The lower the rejection thresholds, the higher the number of trial attempts required to obtain enough data for averaging. Filtering is the process where undesired signals are removed or reduced from each trial record. This can be done during the actual recording (online) or after the record has been made (offline). Additionally, the filtering can be performed with analog circuits or with digital signal processing techniques. The most common configuration is a bandpass filter (or a combination of a low-pass and high-pass filterss. Literatur) that allows alternating current signals with a frequency between the upper and lower cutoffs to pass unattenutated; signals that are outside the bandpass domain are attenuated or eliminated. The upper and lower bounds of the filter are generally set between 300 and 1,500 hertz (Hz)s. Literatur ^,​ s. Literatur as most of the energy from BAER is found between 400 and 1,400 Hz.s. Literatur These settings should reduce the impact that 50- to 60-Hz signals have on the record and processed waveforms. Additional “notch” filters can be used to reduce the impact of 60-Hz electric line interference, though some authors caution against the use of these filters as there is the potential for distortion of the recorded waveforms.s. Literatur ^,​ s. Literatur

The stimuli utilized to elicit AEPs range from clicks to tone pips to pure tones. Intraoperative AEP monitoring typically utilize clicks.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur Clicks, by virtue of their physical properties, stimulate large regions of the cochlea and give robust responses,s. Literatur which is important as large responses make averaging more efficient. The rates of stimulation are typically between about 10 and 30 Hz; other reports have used higher rates when indicated.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur Higher rates of stimulation allow for quicker accumulation of enough trials to generate reliable responses but may reduce the amplitude of the responses.s. Literatur ^,​ s. Literatur ^,​ s. Literatur If the recorded waveforms are small, then lower stimulation rates are employed to reduce this effect.s. Literatur Additionally, stimulus presentation rates should not be at values that are even divisors of 60 Hz to avoid confounding the responses with 60-Hz electrical line interference.s. Literatur ^,​ s. Literatur Click stimuli can be one polarity (condensation or rarefaction) or alternating polarity. Alternating polarity stimuli can be used to average out any contribution of the cochlear microphonic, if this signal is not desired.s. Literatur ^,​ s. Literatur Lastly, the stimulus intensity is typically high (>70 db HL) as this can provide large and reproducible waveforms. Stimulation intensities as high as 95 dB HL can be required if the pure tone thresholds are significantly depressed. However, higher intensity stimulation can result in acoustic crossover and possible responses from the contralateral ear and can require a masking stimulus to the nontest ear.s. Literatur ^,​ s. Literatur

A variety of equipment and supplies are utilized in the generation and recordings of AEPs. In general, these can be broken down into three components: a stimulus source and delivery method, recording electrodes, and a signal recording and processing computer. Acoustic stimuli can be generated by a receiver placed in the external ear canal or located remotely from the external ear but delivered by insert earphones to the external ear.s. Literatur ^,​ s. Literatur ^,​ s. Literatur The latter configuration has some distinct advantages: physically separated receivers allow for better electrical shielding and a time delay for the stimulus to travel to the external ear canal.s. Literatur ^,​ s. Literatur ^,​ s. Literatur Shielding helps reduce the amount of electrical artifact created by the receiver as it generates the clicks (AKA stimulus artifact). The time delay due to the distance that the stimulus must travel from the receiver through the tubing to the inert earphone additionally helps reduce the impact of any stimulus artifact. This is possible because the time delay provides temporal isolation of the AEPs from any stimulus artifact.s. Literatur ^,​ s. Literatur Recording electrode can be either skin surface contacts or subcutaneous needles, as outlined above. For CNAP recording, an additional electrode needs to be placed on or near the cochlear nerve. This occurs after the craniotomy has been performed. Various electrodes have been used and designed for CNAP, and examples include a simple wire, a wire with a ball-tipped end, bipolar wire electrodes, flat contacts, and a specialized “C”-shaped electrode.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur Some of these examples are fabricated each time by the surgeon or institution; others are commercially available.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur The electrodes are connected to a recording device, typically an amplifier attached to a computer. In contemporary AEP systems, the computer drives the generation of the acoustic signals, recording and processing of any responses, and display of the waveforms for interpretation. These systems allow for the control of a variety of stimulus types and rates, filtering options, artifact rejection criteria, and recording parameters. Lastly, a skilled technician or neurophysiologist is required to assess the readout from the AEPs recording and determine if a change in the response pattern has occurred.

28.4 BAER versus CNAP

The overall goal of AEP monitoring of the eighth nerve during surgical management of VS is to identity manipulations that are resulting in physiologic changes likely to produce a decrement in residual hearing and to predict hearing function at the termination of the case.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur The ideal monitoring system would 1) provide clear and stable electrophysiologic responses from the structures at surgical risk; 2) responses would be measured in real time and changes in these responses would be quickly, reliably, and objectively identified; 3) changes in these responses would represent reversible dysfunctions amenable to intervention; 4) changes would reflect the exact structure that is being harmed; and 5) the responses would reliably predict postoperative hearing outcomes. At this time, no system or strategy fulfills fully all or even most of these criteria. However, both BAER and CNAP can provide the meaningful estimates of residual physiologic function of the auditory system. Each has its advantages and disadvantages. Table 28‑1 gives a general comparison of BAER and CNAP monitoring techniques.

BAER is a far-field recording technique, meaning that the recording electrodes are located at a distance from the neural generators of the AEP signals.s. Literatur ^,​ s. Literatur ^,​ s. Literatur Advantages of BAER include the familiarity of BAER waveforms to most neurotologists and neurophysiologists, the capacity to obtain preoperative measurements, consistent placement of electrodes relative to the neural generators, and a generally stable recording environment as the electrodes are located away from the operative field. BAER is a standard technique used by many neurotologists, neurologists, audiologists, and researchers and thus the waveform morphologies, latencies, and interpretations are familiar (Fig. 28‑2 a).s. Literatur Preoperative recording can be compared with intraoperative findings to assess for changes or technical issues at any time in the case. The stability of the electrode placement allows for more reliable assessment of changes. When the relative position of an electrode changes, this will alter the morphology of the waves of interest and may confound the ability to determine if a change is due to a surgical manipulation. There is one main and significant disadvantage to BAER: the relatively small size of the potentials as measured by the technique (nanovolts to microvolts). This results in a need for many averaged trials to reduce the impact of other confounding signals. It is not uncommon for several thousand trials to be required to discern BAER waveforms. This can take a considerable amount of time, frequently several minutes, and reduces the temporal accuracy of BAER monitoring to alert the surgeon of a potentially injurious maneuver prior to irreversible damage.

CNAP is a near-field recording technique, meaning that one of the electrodes is located very close to the neural generators of the AEP signal.s. Literatur This technique was introduced as a means to overcome the disadvantages of BAER, specifically the small measured amplitude, which requires significant averaging to reduce contaminating signals.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur The CNAP response takes a triphasic waveform with a vertex positive initial wave and then larger negative wave, followed by a second positive wave (P1N1P2 configuration).s. Literatur The P1 wave can be very small and there is sometimes a fourth negative wave (N2) resulting in the N1P2N2 and P1N1P2N2 configurations.s. Literatur Fig. 28‑2 b demonstrates the CNAP waveform we typically identify at the University of Iowa. There is some disagreement in the published literature regarding the naming convention as some authors begin naming with the first negative wave; this is the convention used at the University of Iowa (Fig. 28‑2 b).s. Literatur ^,​ s. Literatur ^,​ s. Literatur The source generators and latencies of these responses change depending on where the inactive electrode is placed: distal CN in the internal auditory canal (IAC) or proximal CN in the medial cerebellopontine angle at or near the brainstem. In the initial description of the intraoperative CNAP, Møller and Jannetta described the latency of the CNAP matching with the wave II of the BAER, indicating a proximal CN source.s. Literatur Several authors have described changes in the waveform morphology depending on the exact placement of the inactive electrode.s. Literatur ^,​ s. Literatur ^,​ s. Literatur

When used in the middle fossa approach, the electrode is placed near the distal CN,s. Literatur ^,​ s. Literatur whereas when utilized in the retrosigmoid or suboccipital approaches the electrode is placed near the proximal CN at the brainstem, in some cases near the foramen of Luschka.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur The main and significant advantage of CNAP is the relatively large evoked potentials. This allows for rapid signal identification, as the number of averaged trials can be low, resulting in near-real-time readout of auditory system physiology.s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur ^,​ s. Literatur This near-real-time assessment may allow for the identification of an injurious surgical maneuver in time to alter the maneuver or to intervene and potentially preserve residual physiologic function. We have found that with CNAP recordings, we can identify reliable waveforms in as few as 100 trials in many patients, which is in agreement with the published literature.s. Literatur ^,​ s. Literatur Other authors have reported CNAPs requiring 100 to 300 averages to discern a waveform.s. Literatur ^,​ s. Literatur ^,​ s. Literatur There are several disadvantages to CNAP: there is no method to measure CNAPs preoperatively until the tumor is exposed, the electrode placement is generally less stable than BAER, the waveforms are less familiar to most neurotologists and neurophysiologists, and there is not a currently available standard electrode. The inability to obtain preoperative CNAP measurements is not significant impediment since, if a BAER can be recorded, it is highly likely that a CNAP can be obtaineds. Literatur (unless the residual auditory physiology has been compromised by the surgical approach; see below). Many reports demonstrated CNAP responses in the setting of absent BAER either preoperatively or postoperatively.s. Literatur ^,​ s. Literatur Electrode placement and stability are significant issues. Regardless of which surgical approach is undertaken, the electrode and its wire are invariably in the surgical field, making movement or complete dislodgement possible during surgical maneuvers. With movement of the electrode, the morphology and possibly the polarity of the waves can be altered even if the physiology of the generators has not changed.s. Literatur ^,​ s. Literatur This is discussed below but it should be evident that nonphysiologic changes in the signals will make interpretation less reliable. Depending on the exact location of the electrode, different waves of the response will be accentuated or diminished. If the electrode is placed in the IAC, then wave I will be large with smaller later waves. On the other hand, electrode placement at the root entry zone will accentuate wave III. Thus, the morphology is less stable, predictable, and familiar to most neurotologists and neurophysiologists, making interpretation more difficult.

CNAP is the preferred method for intraoperative eighth nerve monitoring at the University of Iowa, despite its shortcomings, as it can provide near-real-time estimates of the physiological impact of surgical manipulations. Additionally, it is quick and easy to switch back to BAER if CNAP is not stable or reliable enough to be of practical use during the case.