Synonyms

Auditory Brainstem Responses (ABR; has been suggested as the standard abbreviation; Davis 1979); Brainstem Auditory Evoked Response (BAER – used consistently in neurology; Hall 2007); Brain Stem Auditory Evoked Potentials; Auditory Brainstem Potentials; Auditory Brainstem Evoked Potentials

Definition

ABRs are far-field evoked potentials, generated by the brainstem auditory pathways. They are the earliest auditory evoked potentials that can be extracted from the electroencephalogram, recorded from the scalp with electrodes placed on the vertex and referenced to the earlobe or mastoid. There are two types of ABRs, the transient-evoked ABRs, normally evoked by high-intensity clicks presented at a high rate and the steady-state frequency following potentials (sustained phase-locked responses), which can be evoked by tones of a single polarity and phase (Picton 2011; Pratt 2012).

Detailed Description

Transient-Evoked ABRs

Human transient-evoked ABRs were first systematically described by Jewett, Romano and Williston in 1970 (Jewett et al. 1970; Jewett and Williston 1971). They consist of a series of six or seven positive waves, which are generated in the first 10 ms after stimulus onset and are spaced at intervals of about 1 ms (Pratt 2012). They are labeled by the Roman numerals I to VII, as introduced by Jewett and Williston (1971). The amplitude of ABR components are in the order of tenths of a microvolt (Pratt 2012), smaller than 0.5 μv (Møller 2006), and the largest component is wave V (Picton 2011). Some authors prefer to plot the ABR with the positive Waves as an upward deflection, whereas others display the ABR with the positive waves as a downward deflection (Møller 2006).

The auditory stimulus used to elicit ABR is normally a brief duration click, presented at an intensity of 70 dB nHL (normalized Hearing Level; normal Hearing Level. Sound intensity is objectively measured in dB SPL (Sound Pressure Level), defined as 0.0002 dyn/cm2 or 20 microPascals (20 μPa). However, a common convention in clinical settings is to define intensity with a biological or behavioral reference, in dB relative to the hearing threshold level of a particular stimulus for a group of ten to fifteen normal-hearing young adults, which is then indicated in dB nHL (normal Hearing Level; Hall 2007).) and at a rate close to 10/s (Picton 2011) or 20/s (Hall 2007). When measuring ABR, the stimulus polarity must also be taken into account (Hall 2007). The ABR can be best recorded with an electrode at the vertex, referenced to an electrode in the vicinity of the stimulated ear (mastoid or earlobe; Pratt 2012). Generally, 2,000 trials are averaged for showing the ABR (Picton 2011). For extracting the ABR from the auditory evoked potential, a frequency filter is necessary. The ABR has a high frequency and therefore a low-pass filter of between 2,000 and 3,000 Hz and a high-pass filter of between 5 and 30 Hz are typically applied (Picton 2011). The sampling rate of the electroencephalographic signal should not be lower than 20 kHz (Pratt 2012).

Waves I and II are generated by the auditory nerve (Gelfand 2010). More precisely, Wave I originates from the ipsilateral distal eight nerve and Wave II from the ipsilateral proximal eight nerve (Stone et al. 2009) in the vicinity of its entry into the brainstem (Pratt 2012). Wave III has been attributed to the ipsilateral cochlear nucleus and the superior olivary complex (Stone et al. 2009). Wave IV is not always identified in human subjects and is usually partially merged with Wave V (Pratt 2012). It has bilateral multiple brainstem origins (Stone et al. 2009). The sources of Wave V are the contralateral distal lateral lemniscus and inferior colliculus (Stone et al. 2009). Wave VI has been attributed to the medial geniculate body (Gelfand 2010). However, there are divergent opinions as to the exact origins of the ABR waves and it is known that multiple generators in the auditory brainstem contribute to the waves beyond Wave II (Gelfand 2010).

Non-pathological factors which affect the transient-evoked ARBs are amongst others the subject’s age, body temperature, and gender. Moreover, stimulus factors like frequency composition, intensity, presentation rate, and envelope have an influence (Pratt 2012). High-frequency stimuli and stimuli with a short rise time evoke ABRs with larger amplitudes and shorter latencies (Pratt 2012). Also, with increasing stimulus intensity, the ABR peak latency shortens (Pratt 2012). Anesthetics have little or no effect on the ABR (Picton 2011). Regarding attention, the general consensus is that the ABR is not susceptible to whether the evoking stimulus is attended or ignored (Picton 2011). Even during sleep the ABR does not change significantly (Picton 2011).

Fields of clinical application of ABR recordings are, amongst others, newborn infant auditory screening, the estimation of auditory sensitivity in infants and difficult-to-test children, including frequency specific information at audiometric frequencies, neurodiagnosis of eighth nerve or auditory brainstem dysfunction, monitoring eighth nerve and auditory brainstem status intraoperatively during surgery potentially affecting the auditory system and the diagnosis of auditory neuropathy (Hall 2007).

Frequency-Following ABRs

The frequency-following response (FFR) is a sustained evoked potential that is phase-locked to the individual cycles of the stimulus waveform and/or to the periodicity in the envelope of an acoustic stimulus. The scalp-recorded FFR can be evoked by frequencies of up to 1,500–2,000 Hz (Picton 2011; Gelfand 2010). According to the general consensus, it reflects synchronized activity of neural elements in multiple nuclei of the brainstem, primarily the rostral brainstem (Starr and Don 1988; Krishnan 2006). This assumption is confirmed by the typically observed delay of 4–10 ms between the phase of the response relative to the stimulus phase (Starr and Don 1988; Skoe and Kraus 2010). The recording protocol is essentially similar to the transient evoked ABR recording with a few modifications, e.g. typically a longer rise time is used to minimize the onset response (Krishnan 2006). Headphone transducers should be magnetically shielded or placed some distance from the recording sites by using insert earphones with long tubes to minimize stimulus artifacts which otherwise overlap with the FFR response (Skoe and Kraus 2010). Further, the neural FFR should be separated from the cochlear microphonic (CM) – a receptor potential generated by the hair cells – with an appreciably shorter delay (<2 ms). To minimize the contribution of stimulus artifacts and CM, a contralateral recording technique and/or adding responses to stimuli of reversed polarities are effective, even though the later can also diminish the FFR amplitude and leads to a doubling of the FFR frequency (Krishnan 2006). Steady-state FFR is commonly analysed by performing a Fast Fourier Transformation of the time domain response. For FFR in response to time-varying complex stimuli, time-frequency analysis or short-term autocorrelation algorithms can be applied to extract measures such as strength and accuracy of the periodicity tracking (Skoe and Kraus 2010; Krishnan 2006).

Due to its high threshold of 30–40 dB nHL (Picton 2011), the clinical use of FFR has been considered marginal. In recent years there is renewed interest in the FFR since it carries information about the encoding of complex sounds, such as speech and music, and can give insight into the processing of pitch-relevant information and binaural cues at the level of the brainstem. Applied to clinical and healthy populations research on FFR can yield information on how hearing impairment, experience, and training can alter the neural representation of complex sounds as measured at the brainstem level (Krishnan 2006). New studies widely confirm the short- and long-term plasticity of the brainstem FFR and its relation to learning impairments, problems of hearing in noise, or reading difficulties (Skoe and Kraus 2010).