The auditory brainstem response (ABR)

INTRODUCTION

The auditory brainstem response (ABR) represents the far-field potentials recorded from the scalp, and comprises five or more waves generated from the auditory pathway up to the level of the inferior colliculus. The ABR occurs within 10 ms of a click or a brief tone-burst stimulus (see Fig. 6.1). It was originally described by Sohmer & Feinmesser (1967) and independently by Jewett & Williston (1971). As an audiological and a diagnostic test, ABR is firmly established in clinical practice. Its value is most evident in audiometric threshold estimation and in neuro-otological and neurological diagnosis.

Fig. 6.1 Normal ABRs recorded from left and right sides with ipsilateral click stimulation. The peaks and main diagnostic parameters are shown.

GENERATORS OF THE ABR

The extensive literature on the origins of ABR in laboratory animals and in man has been thoroughly reviewed by Buchwald (1983). The early interpretations of the likely origins of ABR have been gained from experimental lesions in animals (Buchwald & Huang 1975), and from human brain structures during operations and in pathological conditions (Moller & Jannetta 1985).

Determining the origins of ABR in laboratory animals and then extrapolating the data to interpret human ABR is not easy. The cochlear nerve is much longer in man than in small mammals (Lang 1981), the volumes of the nuclei differ – for example, the superior olivary complex is smaller in primates than in other mammals (Moore & Moore 1971) – and, in general, the organization of the human auditory system differs in quality from that of the small mammal.

However, the ABR morphology and responsiveness to various experimental manipulations in experimental animals are similar to those in man (Huang & Buchwald 1978, Stockard et al 1978, Don et al 1977).

Apart from peak I, which in both small mammals and man is generated from the distal part of the peripheral nerve, the subsequent peaks are generated from different anatomical structures according to species. For example, wave IV in the cat appears to be analogous to wave V in man (Allen & Starr 1978, Huang 1980).

In man, wave I originates in the distal part of the cochlear nerve, and radiates as a surface negative wave to periauricular sites (Buchwald 1983). The peak latencies of wave I recorded from a surface electrode and from a needle electrode on the promontory have similar values and reflect the gross neural activity of the cochlear nerve (Sohmer & Feinmesser 1967, Abramovich & Billings 1981).

It was thought that wave II originates mainly in the cochlear nucleus (Buchwald 1983). However, intracranial recordings in man suggest that the peak of wave II originates mainly from the proximal part of the cochlear nerve, but with some contributions from the cochlear nucleus (Hashimoto et al 1981, Moller & Jannetta 1985).

The origins of the subsequent waves generated in the brainstem are not clear. It is now accepted that the sequence of the waves does not correspond to simple rostral propagation of the excitation. Most likely, the various orientations of vectorial fields generated in several nuclei overlap, and the studies in man using intracranial recordings support this (Hashimoto et al 1981, Moller & Jannetta 1983, 1985). Peaks III, IV, V, and the subsequent VI and VII derive contributions from more than one anatomical location. Also, each nucleus may contribute to more than one wave in the ABR. Topographic analysis of the auditory-evoked potentials using multichannel recordings, combined with theoretic dipole modelling, suggests that wave III is mainly generated in the region ranging from the ipsilateral cochlear nuclei to the contralateral superior olivary complex, and that the IV–V complex seems to originate from the ipsilateral and contralateral lemniscal structures (Scherg & von Cramon 1985).

One of the most constant morphologic features of the ABR is a vertex–positive peak V, which is followed by a prominent negative potential at the vertex with a latency of 7 ms. This has been named far-field potential 7 (FFP7) by Terkildsen et al (1974). A similar response, especially to low-frequency tone bursts, following wave V at 10 ms is the slow, negative SN10 response shown in Figure 6.2 (Davis & Hirsh 1979). The SN10 is consistent during muscle paralysis, confirming its neural origin (Fria et al 1984).

Fig. 6.2 The SN10 response recorded using a band-pass filter of 40–3000 Hz. (After Davis H 1984 with permission.)

Each region of the cochlea initiates an ABR from the ascending pathways. Thus, for a click stimulus which excites all of the cochlea, the response that is recorded from a surface electrode is the sum of the responses from all the regions in the cochlea. Each of these ‘elemental’ responses exhibits the ‘travelling wave’ delay before its initiation. Thus, the overall ABR is a compound response that represents the sum of the elemental responses. Because of the delays involved, some peaks of the elemental responses are cancelled and some are superimposed. Techniques such as the ‘derived-response’ method can extract frequency-specific responses from the compound ABR (see p. 37).

Williston et al (1981) and Pratt et al (1984), using three orthogonal electrode pairs, have recorded ABR waveforms and produced three-dimensional representations in voltage-time and voltage-space. When this was done, certain segments of the plot lay in a single plane (Fig. 6.3). Furthermore, the orientation of the plane can be altered by disorders of the auditory system. Current research is refining the interpretation of the data and investigating the effects of various pathologies.