Elsevier

Hearing Research

Volume 138, Issues 1–2, December 1999, Pages 115-132
Hearing Research

Basic properties of the sound-evoked post-auricular muscle response (PAMR)

https://doi.org/10.1016/S0378-5955(99)00159-8Get rights and content

Abstract

One objective electrophysiological test for deafness involves presenting a brief acoustic stimulus to a subject and measuring the electrical activity evoked in the muscle located just behind the ear (the post-auricular muscle or PAM). Although this electrical response has been known for many years, it has been ignored by most clinicians and frequently misreported in the literature. This paper presents the fundamental properties of the PAM electrical response (the PAMR) and examines ways in which its measurement can be improved by altering the standard electrode position and filtering. The response consists of a simple bipolar compound action potential with a first peak latency of between 12.5 and 15 ms, depending on the stimulus intensity and PAM muscle tone. The largest recordings can be made with an active electrode over the PAM and with the reference electrode on the dorsal surface of the pinna. It can be obtained with click and tone-burst stimuli within 20 dB of the subjective detection threshold, can be evoked with tone-bursts between 500 Hz and 16 kHz and grows either linearly with the click level or approximately exponentially with the tone-burst level, reaching a maximum of as large as 250 μV pp in some subjects. It has a frequency spectrum mostly between 25 and 200 Hz. The response is often visible in raw recordings, with as few as 20 averages required for obtaining a stable waveform. There is very little amplitude and latency difference in stimulating the ear on the same side or opposite side to the recording electrodes and the binaurally evoked response is similar to the simple arithmetic sum of the waveforms obtained with monaural stimulation. The response latency and duration are longer in very young infants, but reach adult values by 12 months of age. In a companion paper, we describe a method of enhancing the PAMR using lateral eye movement (Patuzzi and O’Beirne, 1999a).

Introduction

The post-auricular muscle response (PAMR) is a large sound-evoked muscle action potential that can be measured on the skin surface over the muscle behind the ear. It can be triggered by rapid onset acoustic stimuli such as clicks or tone-bursts and, with the active electrode over the PAM, consists of two peaks (inverted in Fig. 1 according to normal clinical practice): a negative-going peak occurring between 12.5 and 15 ms and a positive-going peak between 15 and 18 ms. This paper describes some of the basic features of the PAMR, including the optimal electrode placement for its recording, its spectral content, changes in its latency and amplitude with the level of click and tone-burst stimuli and other properties relevant to its use in screening for deafness.

It is clear from previous work that the cochlea is the receptor organ driving the PAMR, since it can be obtained from subjects with abnormal vestibular function but normal hearing, but is absent in deaf subjects with normal vestibular function (Yoshie and Okudaira, 1969, Gibson, 1975). This point is salient, since it is now understood that high level clicks (>85 dB sensation level (SL)) can stimulate the sacculus and produce measurable (inhibitory) responses in some postural muscles (Didier and Cazals, 1989, Colebatch et al., 1994). It is also clear that the PAMR is of muscular, rather than neural, origin. Bickford et al. (1964) have found that its amplitude could be enhanced or abolished by contraction or relaxation of the PAM and that a local anaesthetic block of the post-auricular branch of the facial nerve abolished the response unilaterally.

The PAMR is often much larger than the more commonly recorded auditory brainstem response (ABR) (peaks I, III and V of the ABR in the first 10 ms of the waveform of Fig. 1) and its amplitude alters with the muscle tone of the PAM. Four average waveforms are presented in Fig. 1 (n=2048, bandwidth 10 Hz–5 kHz): two were obtained with the eyes facing forward, and consequently, the PAM was relaxed and the PAMR was small, and two others were obtained with the eyes hard to the right, which increased the PAM’s muscle tone and the PAMR amplitude (see Patuzzi and O’Beirne, 1999a). Except where stated otherwise, all results in this report were obtained with the eyes facing hard towards the PAM recording electrodes, so that the PAMR was large and stable. In the example of Fig. 1 where the PAMR was enhanced by eye rotation, the PAMR was 30 μV pp while the ABR in the same trace was less than 1 μV pp, despite the sub-optimal positioning of the active electrode on the forehead and the indifferent electrode on the temporal bone (see later). In fact, the PAMR is often so large that it can be seen clearly in the raw trace, as illustrated in Fig. 1B, where five raw (unaveraged) waveforms are presented. In this case, the stimulus was a click at 50 dB SL and the active electrode was positioned directly over the PAM, while the reference electrode was placed on the pinna. The higher signal to noise ratio of the PAMR relative to the ABR means that less amplification is required to observe it and much less averaging is needed to produce a stable averaged waveform. In the case of Fig. 1A, 2048 samples were averaged to stabilise the ABR section of the waveform, but only 20 are necessary to produce a stable PAMR trace, when electrode placement is optimised.

If the PAMR is so easily obtained, why is it not more commonly used in place of the ABR as a screening test? One of the major obstacles to its widespread use as an objective hearing test is its perceived variability across the population and within individuals. The PAMR was described by Picton et al. (1974) as ‘highly variable from subject to subject and even within subjects’. Cody and Bickford (1969) found the response to be absent in at least one ear of 32% of their subjects and absent bilaterally in 7% of their subjects. Because of this variability, they considered it unlikely that the PAMR would have any useful clinical application. Similarly, Suzuki stated that ‘from the audiological point of view, the most serious disadvantage of the PAMR is the inconsistency or variability of its appearance.... Such an individual variability is a very serious problem for applying the response as an index of objective audiometry. However, we should overcome this problem because the response is a very important and easily recordable one...’ (Suzuki, in Bochenek and Bochenek, 1976). As we describe in a companion paper, much of this variability may be due to uncontrolled eye movements in the test population (Patuzzi and O’Beirne, 1999a). We have also investigated the reasons for the absence of the PAMR in one subject in detail (Patuzzi and O’Beirne, 1999a).

In any case, dismissal of the PAMR as a screening test has not been universal. Based on his own research, Gibson (1975) has concluded that tests based on the PAMR appeared to provide an excellent method of assessing the hearing sensitivity of children during a clinic: ‘The advantages of the test are that all manner of children, normally untestable without sedation, can be rapidly screened during the course of the actual clinic’. He felt that the unique advantage of the PAMR was that ‘since it is a muscle response, the tense child difficult to test by other means gives clear responses’. Gibson believed the disadvantages of the test were that (i) click-evoked responses cannot accurately reproduce a pure-tone audiogram (but see our later results), (ii) the judgement of the actual hearing threshold is not as accurate as that obtained using electrocochleography or by cortical-evoked response audiometry in older children1 and (iii) poor responses are obtained with conductive hearing losses (particularly in children with serous otitis media)2. However, due to the speed and ease with which the response can be obtained, Gibson felt that its advantages far outweighed its disadvantages and had used the PAMR as a routine clinical tool at the Hearing and Language Clinic at Guy’s Hospital, London, for a number of years (Douek et al., 1974). In any case, while there may be some subjects who do not possess an easily measured PAMR, the majority (70–80%) of subjects possess the response near subjective threshold (Buffin et al., 1977, Flood et al., 1982, Gibson, 1975).

In this report, we summarise some of the basic properties of the PAMR and outline the conditions that are optimal for recording it. Our report is not intended as a population study, rather, we present detailed data from a few subjects (five adults and two infants) who are typical of more than 30 subjects. In a companion paper (Patuzzi and O’Beirne, 1999a), we describe a method for enhancing the PAMR in most subjects, which may reduce the variability of the response greatly.

Section snippets

Subjects

In this study, five adult and two infant subjects were studied in detail a preliminary detailed study of a small number of subjects representative of the 30 subjects tested so far, before a full population study. The pure-tone audiometric thresholds of the adult subjects were measured using a Diagnostic Audiometer (model TA155) at frequencies between 125 Hz and 8 kHz and were found to be normal. Thresholds were not estimated in the infants, other than by the PAMR itself. All tests complied with

Distribution of the PAMR response

The distribution of the PAMR around the post-auricular area has been studied previously by other researchers (Yoshie and Okudaira, 1969, Picton et al., 1974, Streletz et al., 1977, Buffin et al., 1977). However, few studies have considered the distribution of the potential across the pinna (for example, Streletz et al., 1977). This is important for clinical application of the PAMR, as described later. In the present study, the distribution of the response over the post-auricular area and the

Overview

The PAMR is a very large myogenic response evoked by clicks and tone-bursts. It can be detected for sound stimuli as little as 10 dB above subjective threshold in normal and hearing-impaired subjects (Thornton, 1975b). Despite the fact that it is much larger than the less variable ABR, it has largely been ignored in recent times, partly because of its perceived variability, partly because it has traditionally been obtained with averaging hardware which is also capable of detecting the ABR (if

Conclusions

The PAMR is a large and simply measured auditory response, which can be evoked by clicks or tone-bursts (up to 8 kHz or higher). The response is optimally recorded as the potential difference between the skin overlying the PAM and the rear of the pinna, with a bandwidth from 10 to 300 Hz. This electrode positioning also reduces background electrical interference and muscle activity and eliminates blink artefacts. Under these conditions, the background muscle noise comes mostly from the PAM

References (31)

  • W. Bochenek et al.

    Postauricular (12 msec latency) responses to acoustic stimuli in patients with peripheral, facial nerve palsy

    Acta Otolaryngol.

    (1976)
  • J.T. Buffin et al.

    The post-auricular muscle response in children

    J. Laryngol. Otol.

    (1977)
  • R.E. Clifford-Jones et al.

    Crossed acoustic response combined with visual and somatosensory evoked responses in the diagnosis of multiple sclerosis

    J. Neurol. Neurosurg. Psychiatr.

    (1979)
  • D.T.R. Cody et al.

    Averaged evoked myogenic responses in normal man

    Laryngoscope

    (1969)
  • J.G. Colebatch et al.

    Myogenic potentials generated by a click-evoked vestibulocollic reflex

    J. Neurol. Neurosurg. Psychiatr.

    (1994)
  • Cited by (88)

    View all citing articles on Scopus
    View full text