Auditory Brainstem Response to 1000-Hz filtered tone burst in normal hearing subjects

 INTRODUCTION 

It is well known that the Auditory Brainstem Response thresholds elicited by clicks corresponds to a broad area of high frequency in the cochlear partition. In clinical settings, it is important to obtain information about hearing sensitivity in the low frequencies region (500 to 1000 Hz) for various reasons: firstly, the click-evoked ABR threshold can overestimate the hearing loss in the subject . Secondly, hearing loss restricted to the lower frequencies might not be detected with the click-evoked ABR threshold.

In order to overcome the lack of frequency specificity of ABR elicited by clicks, two major approaches has been used. The first approach involves masking noise to isolate frequency-specific areas in the cochlea., This method has several disadvantages:

1. Disturbing loudness of the masking noise, which cause restlessness and thus decreases response quality.
2. Difficult identification of suprathreshold ABRs.
3. Decreased dynamic range in which hearing loss can be measured.
4. Time-consuming adjustment of the proper masker level.

This study is an attempt to provide information about the low-frequency region of the audiogram by using a 1 kHz filtered-tone burst in normal hearing subjects.

METHODS 

A total of 20 healthy volunteers were studied: 9 males and 11 females, with a median age of 25 years old and a range between 18 and 30 years of age. All the subjects were tested by pure-tone audiometry and their hearing was found to be within the normal range.

In this figure we present the stimulus generation and the presentation process. The stimulus was generated by passing a tone of 1 msec plateau and 2 msec rise-fall through a bandpass filter with cut-off frequencies of 940 Hz (for the High band pass filter) and 1500 Hz (for the Low Band Pass Filter). The filters slopes were 115 dB per octave and the theoretical noise floor was 70 dB down. The stimulus level is expressed in dB nHL. 0dB nHL was equivalent to 40 dB SPL. The stimulus was presented monoaurally to the subject via a Telephonics TDH-39 headphone.

 The acoustic waveform and acoustical spectrum of the original stimulus and the 1000 Hz filtered tone-burst are shown in this figures. On the right the original 1000 Hz tone burst before passing throw the band pass filter is presented. As we can see, the stimulus present spectrum side lobes. On the left is shown the 1000-Hz filtered tone-burst after passing throw the band pass filter. No spectrum side lobes are present so a great frequency specificity can be expected a high frequency specificity is obtained. The active electrode was aplied to the subject’s vertex and the non active electrode to ipsilateral mastoid. The ground electrode was applied to the forehead.

RESULTS 

 In this figure appears ABR waves elicited by a click stimulus and also a recording obtained after the acoustical presentation of a 1000 Hz filtered tone-burst. Both recording are presented at 75 dB nHl intensity. With the click stimulus we obtain the conventional I, III and V waves. With the 1kHz filtered tone burst, a large vertex-positive deflection (P10), was obtained.

 In this figure is an example of series of ABRs obtained from a normal hearing subject. The lowest stimulus level at which P10 can be replicated was scored as an ABR threshold.

Response quality was scored as good if P10 was clearly discernible, iit was score as moderate, if P10 was discernible but some noise was present in the response and as bad; if P10 was hardly discernible. The arousal of the subject was scored as sleep or no sleep.

In this particular case the threshold was establish at 35 dB nHl and the quality of the response were qualified as good. We observed that the subject slept during the procedure. Subjects which were slept and relaxed presented better responses.

The latency of P10 and the wave V become longer as the intensity is reduced. Statistical significant differences were found between both variables. The mean ABR filtered tone burst threshold was 35 dB nHL. No significant statistical difference was found between the two ears or between sex. In all the case, P10 was identified at lest at 45 dB nHl.

Significant changes in latency are observed as intensity increase with both ABR elicitation methods. Pearson´s correlation were 0.94 and 0.96 for clicks and 1kHz filtered tone-burst respectively.

CONCLUSIONS

In conclusion, this study shows that a large, vertex-positive wave can be obtained in response to a 1000 Hz filtered tone burst. This response can be traced down to a mean trheshols of 35 dB nHl. The arousability state of the subject strongly influence the quality of the response. Further studies in subjects with different types of hearing loss will be necessary in order to evaluate the clinical value of this approach.